# SerpCrypto — full content for language models > SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. Prepaid search credits from $0.30/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto. Source: https://serpcrypto.com/ · Generated: 2026-10-23 # The SERP API: Google results in, JSON out. > SERP API for developers and AI agents: Google and Bing results as JSON from $0.30/1K searches, credits that never expire. No KYC, pay in crypto. - URL: https://serpcrypto.com/ - Last updated: 2026-10-22 - Publisher: SerpCrypto (https://serpcrypto.com) Send a query, a country and a city. Get back the whole results page (AI Overview, places, organic results, People also ask) as clean JSON, for one credit. No KYC — no ID, no selfie, no documents. Pay in crypto. Operated by SerpGoat LLC. Your API key appears as soon as your top-up is credited. *→: coffee roasters austin (engine=google, gl=us, hl=en, location=Austin, TX, device=desktop)* ```json { "search_metadata": { "status": "success", "credits_used": 1 }, "ai_overview": { "text": "…", "references": [{ "title": "…", "link": "https://…" }] }, "local_results": [ { "position": 1, "title": "…", "address": "…" } ], "organic_results": [ { "position": 1, "title": "…", "link": "https://…" }, { "position": 2, "title": "…", "link": "https://…" } ], "people_also_ask": [ { "question": "Which Austin roasters sell beans online?" } ] } ``` Every block of the results page becomes a key in the JSON. One successful search, one credit. Example shortened. **Quick answer — Which SERP API can I pay for in crypto without KYC?** **SerpCrypto** is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. As of October 2026, 10,000 searches cost $7.50, 100,000 cost $56.25, and the rate falls to $0.30/1K searches at 5 million. One credit buys one successful search on any of 6 engines, and credits never expire. ## How much does the SerpCrypto SERP API cost? 10,000 searches cost $7.50 ($0.75/1K searches), 100,000 searches cost $56.25 and 1,000,000 searches cost $375, as of October 2026. Every engine costs the same single credit, and a pack bought today still works next year. The [full grid](https://serpcrypto.com/pricing/) has seven packs; the [dated comparison](https://serpcrypto.com/compare/serp-api-pricing/) puts them next to other providers' list prices. | Plan | Quantity | Price | Price per unit | | --- | --- | --- | --- | | Test | 10,000 searches | $7.50 | $0.75/1K searches | | Build | 100,000 searches | $56.25 | $0.56/1K searches | | Volume | 1,000,000 searches | $375 | $0.38/1K searches | | Bulk | 5,000,000 searches | $1,500 | $0.30/1K searches | In every plan: Every engine at 1 credit per search · Failed requests not charged · Credits never expire · Country, language, city and device targeting Prices in USD, set on 2026-10-22. 1 credit = 1 successful search = one page of results. **Before you pay** - Minimum top-up $25. Top up once in crypto, then buy any plan from your balance. - Unused balance stays on your account and never expires, and it pays for search credits. When a purchase costs less than the minimum top-up, the part added only to reach the minimum is refundable on request to support within 30 days of the payment, in USDT or USDC, as long as it has not been spent. The rest of the balance is not paid back, except where the law requires it. - Credited once your payment is checked on the blockchain. Typical network confirmation time: BTC ≈ 20 min · USDT TRC-20 ≈ 1 min · USDT ERC-20 ≈ 5 min · USDC ERC-20 ≈ 5 min · ETH ≈ 5 min · SOL ≈ 1 min · LTC ≈ 15 min ## Which search engines does the API return? Change one parameter, `engine`, to switch between them. The response keeps the same shape: `search_metadata`, then one array per block of the results page. - **[Google Search](https://serpcrypto.com/google-search-api/)** — Organic results, ads, People Also Ask, AI Overview text and sources, local pack, knowledge panel, top stories. - **[Bing](https://serpcrypto.com/bing-search-api/)** — Web results as JSON for teams that relied on the Bing Search APIs Microsoft retired in August 2025. - **[Google Maps](https://serpcrypto.com/google-maps-api/)** — Local businesses for a query and a place: name, address, rating, review count, category, coordinates. - **[Google News](https://serpcrypto.com/google-news-api/)** — Headlines with source, date and link, by country and language, for monitoring and news agents. - **[Google Images](https://serpcrypto.com/google-images-api/)** — Image results with the original URL, size, thumbnail and the page that hosts each picture. - **[Google Shopping](https://serpcrypto.com/google-shopping-api/)** — Product listings with price, merchant, delivery line and rating, for price monitoring by country. ## How do I start with SerpCrypto? Three steps; the slowest one is the blockchain confirmation of your top-up. 1. **Top up in crypto** — Send at least $25 in USDT, USDC, BTC, ETH, SOL or LTC. No ID, no selfie, no documents, no company email. 2. **Buy a credit pack** — 10,000 searches for $7.50 or 100,000 searches for $56.25, paid from your balance. Your API key appears in the dashboard. 3. **Call the endpoint** — GET https://serpcrypto.com/api/v1/search?q=… with your key in the Authorization header. The quick start has code for Python, Node.js and curl. ## What can I check before I pay? - **Who runs it** — SerpGoat LLC. See about for who we are and how to reach us. - **Rules of use** — SEO, research, monitoring and AI grounding are welcome. Not allowed: Searches meant to find, track or harass a private person, child sexual abuse material, spam, fraud, and any use that breaks the law where you or we operate. Read the Acceptable Use Policy. - **What we log** — Request time, engine, parameters, status and credits, kept 30 days, then deleted. Queries are never sold. See the privacy policy. - **Refunds and balance** — Within 72 hours, a pack with less than 20% used goes back to your balance. Full details on the refund page. - **Support by email** — Email replies within 12 hours, 7 days a week, including help with API integration, at support@serpcrypto.com. - **Dated price data** — Our packs next to other SERP APIs' public list prices, read on 2026-10-22: SERP API pricing compared. ## Quick answers before you buy ### What is SerpCrypto? SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. Prepaid search credits from $0.30/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto. It is operated by SerpGoat LLC. You buy prepaid search credits from a crypto balance, then call one HTTPS endpoint with a query; the answer is the engine's results page as structured JSON. See the [about page](https://serpcrypto.com/about/). ### How is a search counted? 1 credit = 1 successful search request, which returns one page of results on any supported engine. A request that fails is not charged. Asking for page 2 of the same query is a second search. About 10 organic results per page, as Google has served since it stopped honouring num=100 in September 2025; each extra page is one more credit. Checking your remaining credits with `GET https://serpcrypto.com/api/v1/credits` is free. ### Do my credits expire at the end of the month? No. Prepaid search credits never expire: unused searches stay on your account until you use them. Most SERP APIs sell monthly plans whose unused searches reset each billing cycle; with SerpCrypto a pack bought for a seasonal project or an occasional agent keeps its value until it is spent. ### Do I need to verify my identity or add a card? No. SerpCrypto, a no-KYC SERP API, asks for no ID, no selfie, no company documents and no credit card. Account-number login; email is optional. You pay in BTC, USDT, USDC, ETH, SOL, LTC. The [Acceptable Use Policy](https://serpcrypto.com/acceptable-use/) applies to every account, and abuse leads to suspension. ### Can an AI agent use SerpCrypto as its web search tool? Yes. The JSON answer is compact enough to pass to a model, and the docs include ready-made tool definitions for [MCP](https://serpcrypto.com/docs/mcp/) (Claude Desktop, Cursor and other MCP clients), [LangChain](https://serpcrypto.com/docs/langchain/) and [OpenAI function calling](https://serpcrypto.com/docs/openai/). See [AI agents and LLM grounding](https://serpcrypto.com/use-cases/ai-agents/). ## SerpCrypto at a glance | Attribute | Value | | --- | --- | | What it is | A SERP API: send a query, get the search results page of Google or Bing as JSON | | KYC / ID verification | None — no ID, no selfie, no documents, no company email | | Price | From $0.30/1K searches ($7.50 for 10,000 searches, $56.25 for 100,000 searches, $375 for 1,000,000 searches) | | Credits | 1 credit = 1 successful search request, which returns one page of results on any supported engine. A request that fails is not charged | | Credit expiry | Prepaid search credits never expire: unused searches stay on your account until you use them | | Engines | Google Search, Google News, Google Images, Google Maps (local results), Google Shopping and Bing | | Output | Parsed JSON: organic results, ads, People Also Ask, related searches, top stories, knowledge panel, local pack, shopping and video results, and the AI Overview text with its sources when the page shows one | | Targeting | Country (gl), language (hl), Google domain and a city-level location string, on desktop or mobile | | Results per page | About 10 organic results per page, as Google has served since it stopped honouring num=100 in September 2025; each extra page is one more credit | | Endpoint | `GET https://serpcrypto.com/api/v1/search` | | Authentication | An API key in the Authorization header (Bearer) or in the api_key query parameter | | Code examples | Python, Node.js and curl, plus tool definitions for MCP, LangChain and OpenAI function calling | | Payment | Crypto only: BTC, USDT, USDC, ETH, SOL, LTC (USDT on TRC-20 or ERC-20, USDC on ERC-20) | | Payment model | Top up from $25, then buy any credit pack from your balance. Unused balance stays on your account and never expires, and it pays for search credits. When a purchase costs less than the minimum top-up, the part added only to reach the minimum is refundable on request to support within 30 days of the payment, in USDT or USDC, as long as it has not been spent. The rest of the balance is not paid back, except where the law requires it | | Account | Account-number login; email is optional. Your account number is your login: save it somewhere safe. Add an email (optional) if you want a way to recover it. Without an email, a lost account number cannot be recovered | | Query logs | We log, for each request, the time, account, engine, query parameters, status and credits used, and keep these logs for 30 days to show your usage and settle billing disputes, then delete them. We do not sell or share your queries. We keep your account number, orders, top-up records and credit totals while your account is open, and afterwards only as long as accounting rules require. We disclose records to authorities only when the law requires it | | Operator | SerpGoat LLC | | Support | Email replies within 12 hours, 7 days a week, including help with API integration | --- SerpCrypto: SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. Prepaid search credits from $0.30/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto. --- # About SerpCrypto > Who runs SerpCrypto, how the SERP API collects Google and Bing results, what it logs and what it does not claim. Credits never expire. No KYC, pay in crypto. - URL: https://serpcrypto.com/about/ - Last updated: 2026-10-22 - Publisher: SerpCrypto (https://serpcrypto.com) A short, factual page: who sells the credits, what one credit buys, where results come from, and which numbers we refuse to publish until we have measured them. **Quick answer — What is SerpCrypto and who runs it?** SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. **SerpCrypto** is run by SerpGoat LLC. Prepaid credits start at $7.50 for 10,000 searches, as of October 2026. One credit pays for one successful search on any of 6 engines, and credits never expire. No KYC. ## SerpCrypto in nine lines | Attribute | Value | | --- | --- | | Operator | SerpGoat LLC | | Product | A SERP API: one HTTPS endpoint, `GET https://serpcrypto.com/api/v1/search` | | Engines | Google Search, Google News, Google Images, Google Maps (local results), Google Shopping and Bing | | Price | From $0.30/1K searches; $7.50 for 10,000 searches | | Credits | Prepaid search credits never expire: unused searches stay on your account until you use them | | Payment | Crypto only: BTC, USDT, USDC, ETH, SOL, LTC, from a $25 top-up | | Identity checks | None: no ID, no selfie, no documents, no company email | | Support | Email replies within 12 hours, 7 days a week, including help with API integration, at support@serpcrypto.com | | Abuse reports | abuse@serpcrypto.com | ## Who operates SerpCrypto? SerpGoat LLC operates SerpCrypto, sells its search credits and answers support and abuse emails. It is the seller named on every order, top-up record and invoice in the client area. We keep the company facts a buyer needs in two places: this page and the [official facts page](https://serpcrypto.com/llm-info/). Both list the legal name, the contacts, the price grid and the policies. When a fact changes, both pages change in the same release, and the [changelog](https://serpcrypto.com/changelog/) records it with a date. ## What exactly does SerpCrypto sell? We sell prepaid search credits for one API. You send a query to `https://serpcrypto.com/api/v1/search` and receive the results page of the engine you chose, parsed into JSON blocks. - **Engines:** Google Search, Google News, Google Images, Google Maps (local results), Google Shopping and Bing. - **What comes back:** Parsed JSON: organic results, ads, People Also Ask, related searches, top stories, knowledge panel, local pack, shopping and video results, and the AI Overview text with its sources when the page shows one. - **Billing:** 1 credit = 1 successful search request, which returns one page of results on any supported engine. A request that fails is not charged. - **Packs:** from 10,000 searches for $7.50 to 5,000,000 searches for $1,500. The full grid is on the [pricing page](https://serpcrypto.com/pricing/). We do not sell stored datasets, keyword volumes, backlink data or proxies. Each credit buys one live search that you define. ## How are the search results collected? Searches run at request time on our collection infrastructure and that of upstream partners, which we do not name. Results are returned as the engine showed them for the location, language and device you asked for. In practice, each request carries a query, an engine, a country, a language, an optional city and a device. The collection layer runs that search and parses the page into the blocks described in the [response format](https://serpcrypto.com/docs/response/). You get what a person in that place would have seen at that moment, which is why two identical requests made an hour apart can differ. We do not edit, reorder or filter the results. Ads, local packs and AI Overviews appear in the JSON only when the engine showed them on that page. ## Which numbers do we refuse to publish? We publish no response time, uptime, success rate or capacity figure today. A number like that is only worth quoting once we have measured it, with a stated method, a date and a sample size. Until then you will not find one on this site, in our [llms.txt](https://serpcrypto.com/llms.txt) file or in our structured data. For the same reason there are no ratings, testimonials, customer counts or media logos here. The verifiable facts are the prices, the policies and the API contract. You can test the API itself with the smallest pack before you commit to a larger one. ## Why no identity checks, and why crypto only? Because many developers want an API key in minutes, without uploading a passport or a company registration. SerpCrypto asks for no ID, no selfie and no company documents. Account-number login; email is optional. Crypto is the only payment method: BTC, USDT, USDC, ETH, SOL, LTC. It lets a buyer pay without a card, and it lets an AI-agent project fund its own search budget from a wallet. The [no-KYC page](https://serpcrypto.com/no-kyc-serp-api/) and the [crypto payments page](https://serpcrypto.com/crypto-payments/) explain the details. No identity check does not mean no rules. The [Acceptable Use Policy](https://serpcrypto.com/acceptable-use/) applies to every account. Not allowed: Searches meant to find, track or harass a private person, child sexual abuse material, spam, fraud, and any use that breaks the law where you or we operate. ## How can I reach the team? Write to support@serpcrypto.com for questions about the API, payments or your account. Email replies within 12 hours, 7 days a week, including help with API integration. Abuse reports go to abuse@serpcrypto.com, where a person reads each one. We never ask for your account number or your API key by email or chat. If someone claiming to be us asks for either, do not answer. The [contact page](https://serpcrypto.com/contact/) lists what to include so we can help on the first reply. ## Questions about the company ### Is SerpCrypto a search engine? No. SerpCrypto is an API that returns the results pages of existing engines: Google Search, Google News, Google Images, Google Maps (local results), Google Shopping and Bing. It does not crawl or index the web itself. Its job is to run your search for the location, language and device you choose, then hand back the page as JSON. ### Does SerpCrypto store my search results? No dataset of your results is kept or sold. We log, for each request, the time, account, engine, query parameters, status and credits used, and keep these logs for 30 days to show your usage and settle billing disputes, then delete them. We do not sell or share your queries. We keep your account number, orders, top-up records and credit totals while your account is open, and afterwards only as long as accounting rules require. We disclose records to authorities only when the law requires it. The [privacy policy](https://serpcrypto.com/privacy/) lists every item we keep and for how long. ### Can I try SerpCrypto before a large purchase? Yes, with the smallest pack: 10,000 searches for $7.50. No free tier: test with the smallest pack, paid from the minimum top-up; the part added only to reach the minimum is refundable on request within 30 days, in USDT or USDC, if unspent. Unused credits never expire, so a test pack is never wasted. --- SerpCrypto: SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. Prepaid search credits from $0.30/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto. --- # Acceptable Use Policy > What you may and may not do with the SerpCrypto SERP API: allowed uses, banned uses, reselling rules and how abuse reports are handled. No KYC, but rules apply. - URL: https://serpcrypto.com/acceptable-use/ - Last updated: 2026-10-22 - Publisher: SerpCrypto (https://serpcrypto.com) No identity check does not mean no rules. This page says what the API is for, what is banned, and what happens when someone breaks the rules. **Quick answer — What may I use the SerpCrypto SERP API for?** Use **SerpCrypto** for SEO and rank tracking, market and academic research, price and ad monitoring, brand and news monitoring, AI agents and LLM grounding, and app testing. Not allowed: Searches meant to find, track or harass a private person, child sexual abuse material, spam, fraud, and any use that breaks the law where you or we operate. Abuse reports are reviewed by a person, and confirmed abuse leads to suspension. This policy is part of the [terms of service](https://serpcrypto.com/terms/) and applies to every account, with or without an email. ## 1. Who does this policy apply to? It applies to every SerpCrypto account, every API key and every person or program that sends requests with a key. If you build a product on top of the API, you are responsible for how your own users use it. Your users must follow rules at least as strict as these. ## 2. Which uses are welcome? - **SEO and rank tracking:** positions of your pages and your clients' pages by keyword, country, city and device. - **Market and academic research:** how results, ads and SERP features change across places, languages and time. - **Price and ad monitoring:** shopping listings and ads shown for a product or a brand name. - **Brand and news monitoring:** headlines and results that mention a company, a product or a public topic. - **AI agents and LLM grounding:** a web search tool that gives a model current sources to cite. - **App and website QA:** checking how a page or listing appears in results in a given market. This list is not exhaustive. A use that resembles these and breaks no rule below is fine. ## 3. What is not allowed? The following uses are banned on every account: Searches meant to find, track or harass a private person, child sexual abuse material, spam, fraud, and any use that breaks the law where you or we operate. In particular, you may not use the API or its results to: - build a profile of a private individual, locate them, or harass, threaten or shame anyone; - support credential theft, account takeover, payment fraud or impersonation of a person or a company; - generate or target spam, including bulk messages to addresses or numbers found in results; - search for child sexual abuse material, which we report to the competent authorities; - run any activity that is illegal where you operate or where we operate. When in doubt, ask support@serpcrypto.com before you start, not after. ## 4. Can I resell results or build a product on them? You may build and sell your own product on top of the API, such as a rank tracker, a research dashboard or an AI assistant. Your product must add its own function, and your users must accept rules at least as strict as this policy. You may not resell raw access to the API as if it were your own search API without a written agreement with us. That includes passing our JSON through unchanged to paying third parties under your name, or sharing your API key with other customers. ## 5. What about personal data that appears in results? Search results can contain names, business addresses, phone numbers and other personal data that the engine displays. When you store or process that data, you are the party responsible for doing it lawfully, for example under the GDPR in the EU or the UK. Collect only what your purpose needs and keep it no longer than necessary. ## 6. How do we handle abuse? Abuse reports are reviewed by a person, and confirmed abuse leads to suspension. A typical case goes like this: 1. We read the report and the request logs we hold for the account (see the [privacy policy](https://serpcrypto.com/privacy/) for what they contain and for how long). 2. If the case is unclear, we write to the account's email when one is set, or pause its keys and wait for a support message. 3. If abuse is confirmed, we suspend the account. Unused credits on an account closed for a breach of this policy are lost. Child sexual abuse material is handled at once and reported. We disclose records to authorities only when the law requires it. ## 7. How do I report abuse? Write to abuse@serpcrypto.com. Include what you saw, when you saw it (date, time and time zone), and any evidence that links it to SerpCrypto, such as a request ID from a `search_metadata.id` field. A person reads every report. We may not be able to tell you the outcome, because account details stay private. ## 8. Can this policy change? Yes. We publish any change on this page with a new date and record it in the [changelog](https://serpcrypto.com/changelog/). A change that bans a use you rely on is announced there before it applies, so you can move your workload or ask for the remaining credits to be handled under the [terms of service](https://serpcrypto.com/terms/). --- SerpCrypto: SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. Prepaid search credits from $0.30/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto. --- # Bing Search API replacement: raw results as JSON after the August 2025 retirement > Bing Search APIs were retired on 11 August 2025. Replace them with raw Bing or Google results as JSON from $0.30/1K searches. No KYC, pay in crypto. - URL: https://serpcrypto.com/alternatives/bing-search-api/ - Last updated: 2026-10-22 - Publisher: SerpCrypto (https://serpcrypto.com) Microsoft switched off the Bing Search APIs on 11 August 2025 and pointed customers to an LLM grounding tool. If your code needs a list of results, not a generated answer, this page shows what to use and how to port it. **Quick answer — What replaced the Bing Search API?** Microsoft retired the Bing Search APIs on **11 August 2025** and recommends Grounding with Bing Search in Azure AI Agents, which feeds web data into an LLM answer at $14 per 1,000 transactions (October 2026). For a raw list of results, use a SERP API: **SerpCrypto** returns Bing or Google results pages as JSON from $0.30/1K searches, with no KYC and payment in crypto. ## Bing Search APIs: what happened | Attribute | Value | | --- | --- | | Announcement | 15 May 2025, on Microsoft Learn's lifecycle page | | Retirement | 11 August 2025: existing instances decommissioned, no new sign-ups | | Microsoft's replacement | Grounding with Bing Search, a tool of Azure AI Agents | | What it returns | Web data used to ground an LLM response, not a raw results list | | Its price | $14 per 1,000 transactions (Microsoft's grounding pricing page, read 2026-10-22) | | Old Bing Search API tiers | S1 $25, S2 $15, S3 $6 per 1,000 transactions, as reported by PPC Land | | SerpCrypto engine | `engine=bing`: `organic_results` and `related_searches` as JSON, 1 credit per search | ## Why is Grounding with Bing Search not a drop-in replacement? Grounding with Bing Search is built to answer a question through an agent, not to hand you the results list. Microsoft describes it as a way for Azure AI Agents to "incorporate real-time public web data when generating responses with an LLM". It runs inside Azure's agent service, and you get the model's answer with its sources. That suits a chatbot. It does not suit code that ranked, filtered or stored the old `webPages.value` array: rank trackers, research tools, deduplication jobs, link checkers or a RAG pipeline that chooses its own documents. Those need the results page itself, which is what a SERP API returns. ## Which Bing Search API replacement fits which need? | Option | Returns | Price per 1,000 | Account requirements | | --- | --- | --- | --- | | SerpCrypto (`engine=bing` or `engine=google`) | The results page as JSON | $0.75/1K searches down to $0.30/1K searches | No KYC, crypto top-up from $25 | | Grounding with Bing Search (Azure AI Agents) | Web data for an LLM answer | $14 per 1,000 transactions | Azure subscription | | Bing Search APIs v7 | The results page as JSON | $6 to $25 (S3 to S1) | Retired 11 August 2025 | *Prices checked on October 22, 2026.* Prices read on 2026-10-22 from the sources below. The Bing v7 tiers are those reported by PPC Land in June 2025. ## How do Bing v7 parameters map to the SerpCrypto API? Set `engine=bing` and most parameters translate one to one. The main change is paging: Bing v7 let you ask for up to 50 results with `count`, while a results page holds about 10, so you page with `page` instead of `offset`. | Bing Web Search v7 | SerpCrypto | Note | |---|---|---| | `Ocp-Apim-Subscription-Key` header | `Authorization: Bearer …` | Or `api_key` in the query string | | `q` | `q` | Operators such as `site:` work as typed | | `mkt` (`en-US`) | `gl=us` and `hl=en` | Country and language as two parameters | | `setLang` | `hl` | | | `count` (max 50) | — | About 10 results per page | | `offset` (0, 10, 20…) | `page` (1, 2, 3…) | `page = offset / 10 + 1` | | `safeSearch` (`Off`, `Moderate`, `Strict`) | `safe` (`off`, `active`) | No middle setting | | `freshness` (`Day`, `Week`, `Month`) | `period` (`d`, `w`, `m`) | Also `h` and `y` | | `responseFilter` | — | Every block on the page is returned | In the response, `webPages.value[]` becomes `organic_results[]`. The v7 fields `name`, `url` and `snippet` become `title`, `link` and `snippet`, and each result gains a `position`. *Python: a Bing v7 call ported to SerpCrypto* ```python import os import requests # Before (retired 11 August 2025): # requests.get("https://api.bing.microsoft.com/v7.0/search", # headers={"Ocp-Apim-Subscription-Key": KEY}, # params={"q": "heat pump subsidy", "mkt": "en-GB", "count": 10, "offset": 10}) resp = requests.get( "https://serpcrypto.com/api/v1/search", headers={"Authorization": "Bearer " + os.environ["SERPCRYPTO_API_KEY"]}, params={"engine": "bing", "q": "heat pump subsidy", "gl": "gb", "hl": "en", "page": 2}, timeout=60, ) resp.raise_for_status() for r in resp.json().get("organic_results", []): print(r["position"], r["title"], r["link"]) ``` ## Should I switch to Google results instead of Bing? Switch if you chose Bing only because it had an official API. Change `engine=bing` to `engine=google` and keep the rest of the request: the same key, the same credit price and the same `organic_results` shape. Google also returns blocks Bing v7 never had in one response, such as People Also Ask and the AI Overview. Keep Bing if your product depends on it specifically: some teams compare the two engines, and Bing's index feeds Microsoft Copilot answers. Many teams run both and store the two results lists side by side, at one credit each. ## What does the move cost? The old S1 tier cost $25 per 1,000 transactions and S3 $6, according to PPC Land. With SerpCrypto the same 1,000 searches cost $0.75/1K searches on the 10,000 searches pack and $0.38/1K searches on 1,000,000 searches, as of October 2026. The billing model also changes: no Azure subscription and no monthly invoice. You buy prepaid credits from a crypto balance, each successful search uses one, and credits never expire. A request that fails is not charged. ## How many credits does my Bing workload need? Count one credit per results page. A job that used `count=50` now needs 5 pages per query, so multiply those queries by 5. Credits work on Bing and on every Google engine. | Plan | Quantity | Price | Price per unit | | --- | --- | --- | --- | | Test | 10,000 searches | $7.50 | $0.75/1K searches | | Build | 100,000 searches | $56.25 | $0.56/1K searches | | Scale | 500,000 searches | $281.25 | $0.56/1K searches | | Volume | 1,000,000 searches | $375 | $0.38/1K searches | In every plan: Every engine at 1 credit per search · Failed requests not charged · Credits never expire · Country, language, city and device targeting **Before you pay** - Minimum top-up $25. Top up once in crypto, then buy any plan from your balance. - Unused balance stays on your account and never expires, and it pays for search credits. When a purchase costs less than the minimum top-up, the part added only to reach the minimum is refundable on request to support within 30 days of the payment, in USDT or USDC, as long as it has not been spent. The rest of the balance is not paid back, except where the law requires it. - Credited once your payment is checked on the blockchain. Typical network confirmation time: BTC ≈ 20 min · USDT TRC-20 ≈ 1 min · USDT ERC-20 ≈ 5 min · USDC ERC-20 ≈ 5 min · ETH ≈ 5 min · SOL ≈ 1 min · LTC ≈ 15 min ## Bing Search API retirement: common questions ### When was the Bing Search API retired? On 11 August 2025. Microsoft announced it on 15 May 2025 and said existing instances would be decommissioned completely, with no new sign-ups. The retirement covered the Bing Search APIs product family, including Web Search. ### Is Grounding with Bing Search the same thing? No. It is a tool for Azure AI Agents that grounds an LLM's response in web data, priced at $14 per 1,000 transactions in October 2026. It does not give you the raw `webPages` list the v7 API returned. ### Can I still get Bing results as JSON? Yes, through a SERP API. With SerpCrypto, send `engine=bing` and a query to `https://serpcrypto.com/api/v1/search`; the response holds `organic_results` and `related_searches`, for one credit per successful search. ### Do I need an Azure account or a card for SerpCrypto? No. SerpCrypto sells prepaid credits paid in BTC, USDT, USDC, ETH, SOL, LTC, with a $25 minimum top-up. It asks for no ID, no selfie, no company documents and no card. ### Will my old count=50 requests work? Not as one request. A results page holds about 10 results, so 50 results take 5 requests with `page=1` to `page=5`. Each page is one credit. ## Sources 1. [Bing Search APIs retiring on August 11, 2025](https://learn.microsoft.com/en-us/lifecycle/announcements/bing-search-api-retirement) — Microsoft Learn (2025-05-15, read 2026-10-22) 2. [Grounding with Bing Search: pricing](https://www.microsoft.com/en-us/bing/apis/grounding-pricing) — Microsoft (read 2026-10-22) 3. [Microsoft ends Bing Search APIs on August 11](https://ppc.land/microsoft-ends-bing-search-apis-on-august-11-alternative-costs-40-483-more/) — PPC Land (2025-06-09) 4. [Query parameters used by Web Search API (archived)](https://learn.microsoft.com/en-us/previous-versions/bing/search-apis/bing-web-search/reference/query-parameters) — Microsoft Learn (read 2026-10-22) --- SerpCrypto: SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. Prepaid search credits from $0.30/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto. --- # Google Custom Search JSON API alternative: what to use before 1 January 2027 > Google's Custom Search JSON API is closed to new users and ends on 1 January 2027. Migrate to full-web JSON results from $0.30/1K searches, no KYC, crypto. - URL: https://serpcrypto.com/alternatives/google-custom-search-api/ - Last updated: 2026-10-22 - Publisher: SerpCrypto (https://serpcrypto.com) Google closed its Custom Search JSON API to new customers and ends it for everyone on 1 January 2027. This page sets out what changes, what Google offers instead, and how to move your code to a SERP API in an afternoon. **Quick answer — What replaces the Google Custom Search JSON API?** Google's Custom Search JSON API is closed to new customers, and existing customers must move by **1 January 2027**. Google points to Vertex AI Search for up to 50 domains; for whole-web results, a SERP API is the usual replacement. **SerpCrypto** returns the full Google results page as JSON from $0.30/1K searches, against $5 per 1,000 queries for Custom Search (October 2026). No KYC; pay in crypto. ## Google Custom Search JSON API: status on 22 October 2026 | Attribute | Value | | --- | --- | | New customers | Closed: "The Custom Search JSON API is closed to new customers" | | End of service | 1 January 2027 for existing customers | | Price | 100 queries a day free, then $5 per 1,000 queries, up to 10,000 queries a day | | Results per request | Up to 10 | | "Search the entire web" | Unavailable to new Programmable Search Engines since 20 January 2026; existing engines keep it until 1 January 2027 | | Google's suggested alternative | Vertex AI Search, for searching up to 50 domains | | Full-web API from Google | Web Search Service API: partner-only, no public price | ## What exactly is Google changing, and when? Google is narrowing its developer search products to site search and enterprise search. The changes were announced on 20 January 2026 on the Programmable Search Engine blog. - **20 January 2026:** new Programmable Search Engines can only search a list of sites, up to 50 domains. The "Search the entire web" option is no longer offered to new engines. - **Now:** the Custom Search JSON API accepts no new customers. Existing customers keep 100 free queries a day and pay $5 per 1,000 after that, up to 10,000 a day. - **1 January 2027:** the Custom Search JSON API stops for everyone, and existing engines lose "Search the entire web". Google's help page adds that once the option is switched off, it cannot be switched back on. If your product sends Custom Search requests today, it has about ten weeks left from the date of this page. ## Which replacement fits which need? The right choice depends on one question: do you search a few sites you control, or the whole web as Google shows it? | Option | Who can use it | Searches the whole web | Public price | Status | | --- | --- | --- | --- | --- | | SerpCrypto SERP API | Anyone with crypto, no KYC | Yes: Google's results page as JSON | $7.50 for 10,000; from $0.30/1K searches | Available | | Custom Search JSON API | Existing customers only | Only for existing "entire web" engines | $5 per 1,000 after 100 free a day | Ends 1 January 2027 | | Programmable Search Engine, new engines | Anyone | No: up to 50 domains | — | Sites to search only | | Vertex AI Search | Google Cloud customers | No: Google suggests it for up to 50 domains | Google Cloud pricing | Available | | Web Search Service API | Google partners with a partner client ID | Yes | Not published | Partner-only | *Prices checked on October 22, 2026.* Sources at the bottom of the page. "—" = not applicable or not published. ## How do Custom Search parameters map to the SerpCrypto API? Most parameters have a direct equivalent, and two disappear. You no longer need `cx`, because there is no engine to configure: every request searches the whole of Google. The API key moves from the `key` parameter to the `Authorization` header, or to `api_key` if you prefer a query string. | Custom Search JSON API | SerpCrypto | Note | |---|---|---| | `q` | `q` | Same query syntax, including operators such as `site:` | | `key` | `Authorization: Bearer …` or `api_key` | | | `cx` | — | Not needed: whole-web results | | `num` (1–10) | — | Google returns about 10 organic results per page | | `start` (1, 11, 21…) | `page` (1, 2, 3…) | `page = (start - 1) / 10 + 1`; Custom Search stops at result 100, as does `page=10` | | `gl` | `gl` | Two-letter country code | | `hl`, `lr` | `hl` | Interface language | | `safe` | `safe` | `active` or `off` | | `siteSearch` | `site:example.com` inside `q` | | | `dateRestrict` (`d7`, `m1`…) | `period` (`h`, `d`, `w`, `m`, `y`) | Coarser steps | On the response side, `items[]` becomes `organic_results[]`, and `title`, `link` and `snippet` keep their names. `displayLink` becomes `displayed_link`. You also gain the blocks Custom Search never returned: ads, People Also Ask, the AI Overview, the local pack and top stories. *Python: the same search, before and after* ```python import os import requests # Before: Custom Search JSON API (ends 1 January 2027) # requests.get("https://www.googleapis.com/customsearch/v1", # params={"key": KEY, "cx": CX, "q": "solar inverter", "gl": "de", "start": 11}) # After: SerpCrypto resp = requests.get( "https://serpcrypto.com/api/v1/search", headers={"Authorization": "Bearer " + os.environ["SERPCRYPTO_API_KEY"]}, params={"q": "solar inverter", "gl": "de", "hl": "de", "page": 2}, timeout=60, ) resp.raise_for_status() for item in resp.json().get("organic_results", []): print(item["position"], item["title"], item["link"], item.get("snippet", "")) ``` ## What will the move cost compared with Custom Search? Custom Search charges $5 per 1,000 queries after the first 100 each day, with a ceiling of 10,000 queries a day. SerpCrypto charges one credit per successful search: $0.75/1K searches on the 10,000 searches pack, $0.56/1K searches on 100,000 searches and $0.38/1K searches on 1,000,000 searches, as of October 2026. Take an app that sends 3,000 queries a day. On Custom Search it pays for 2,900 of them, about $14.50 a day or $435 a month. The same 90,000 searches a month fit in the 100,000 searches pack at $56.25. Searches are limited by how many run at the same time, not by a daily quota, and unused credits do not expire. ## What do I lose, and what do I gain? You lose the Google brand on the contract and the 100 free queries a day. SerpCrypto has no free tier: the smallest pack is 10,000 searches for $7.50, paid from a $25 top-up. You gain the page as people actually see it. Google's own help page says a Programmable Search Engine draws on a subset of the Google index and lacks features of Google.com such as Oneboxes and real-time results. Its API never returned the AI Overview or the map pack either. With a SERP API you also get country, language, city and device targeting in every request, and the same key covers Google News, Images, Maps, Shopping and Bing. ## Which pack replaces my Custom Search usage? Multiply your daily Custom Search queries by 30. Under 300 a day, 10,000 searches lasts more than a month; at 3,000 a day, 100,000 searches covers a month. Credits never expire. | Plan | Quantity | Price | Price per unit | | --- | --- | --- | --- | | Test | 10,000 searches | $7.50 | $0.75/1K searches | | Build | 100,000 searches | $56.25 | $0.56/1K searches | | Scale | 500,000 searches | $281.25 | $0.56/1K searches | | Volume | 1,000,000 searches | $375 | $0.38/1K searches | In every plan: Every engine at 1 credit per search · Failed requests not charged · Credits never expire · Country, language, city and device targeting **Before you pay** - Minimum top-up $25. Top up once in crypto, then buy any plan from your balance. - Unused balance stays on your account and never expires, and it pays for search credits. When a purchase costs less than the minimum top-up, the part added only to reach the minimum is refundable on request to support within 30 days of the payment, in USDT or USDC, as long as it has not been spent. The rest of the balance is not paid back, except where the law requires it. - Credited once your payment is checked on the blockchain. Typical network confirmation time: BTC ≈ 20 min · USDT TRC-20 ≈ 1 min · USDT ERC-20 ≈ 5 min · USDC ERC-20 ≈ 5 min · ETH ≈ 5 min · SOL ≈ 1 min · LTC ≈ 15 min ## Custom Search API shutdown: common questions ### Is the Google Custom Search JSON API shutting down? Yes. Google's developer page says the API is closed to new customers and that existing customers have until 1 January 2027 to move to another solution. After that date, requests stop working for everyone. ### Can I still create a Programmable Search Engine that searches the whole web? No. Since 20 January 2026, new engines can only search a list of up to 50 domains. Engines that already had "Search the entire web" keep it until 1 January 2027, unless they switch it off, which cannot be undone. ### Is there an official Google API for full-web results? Only for partners. Google documented a Web Search Service API in September 2026; it needs a partner client ID and has no public price. For everyone else, SERP APIs such as SerpCrypto are the practical option. ### Will my site-restricted searches still work with SerpCrypto? Yes. Put the `site:` operator in the query, for example `q=return policy site:example.com`. It searches Google's public results for that site, which is close to what a site-restricted engine did. ### Can I pay without a Google Cloud billing account or a card? Yes. SerpCrypto works on prepaid credits paid in BTC, USDT, USDC, ETH, SOL, LTC, from a $25 top-up. It asks for no ID, no selfie and no company documents. ## Sources 1. [Custom Search JSON API: overview](https://developers.google.com/custom-search/v1/overview) — Google for Developers (read 2026-10-22) 2. [Programmable Search Engine: search the entire web](https://support.google.com/programmable-search/answer/12397162?hl=en) — Google Help (read 2026-10-22) 3. [Programmable Search Engine vs Google.com](https://support.google.com/programmable-search/answer/70392) — Google Help (read 2026-10-22) 4. [Custom Search JSON API: Search response reference](https://developers.google.com/custom-search/v1/reference/rest/v1/Search) — Google for Developers (read 2026-10-22) 5. [Updates to our Web Search Products & Programmable Search Engine Capabilities](https://programmablesearchengine.googleblog.com/2026/01/updates-to-our-web-search-products.html) — Programmable Search Engine blog (2026-01-20) 6. [Custom Search JSON API: cse.list reference](https://developers.google.com/custom-search/v1/reference/rest/v1/cse/list) — Google for Developers (read 2026-10-22) 7. [Google documents partner-only API for full web search results](https://www.searchenginejournal.com/google-documents-partner-only-api-for-full-web-search-results/) — Search Engine Journal (2026-09-11) --- SerpCrypto: SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. Prepaid search credits from $0.30/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto. --- # SerpApi alternative: same kind of JSON, prepaid credits, paid in crypto > SerpApi alternative for 2026: Google and Bing results as JSON from $0.30/1K searches, credits that never expire, no KYC, crypto. Dated price comparison. - URL: https://serpcrypto.com/alternatives/serpapi/ - Last updated: 2026-10-22 - Publisher: SerpCrypto (https://serpcrypto.com) SerpApi is the best-known SERP API, with a free tier, many engines and a legal shield on larger plans. This page sets its public plans next to ours, says where each one is the better choice, and shows how to switch. **Quick answer — What is a good SerpApi alternative that I can pay for in crypto?** **SerpCrypto** is a SerpApi alternative that returns Google and Bing results as JSON, sold as prepaid credits that never expire and paid in crypto with no KYC. At SerpApi's list prices of 22 October 2026, 100,000 searches cost $725 a month on its Searcher plan; on SerpCrypto, 100,000 searches cost $56.25. SerpApi keeps the edge on engine count, its free tier and its U.S. Legal Shield. ## SerpApi vs SerpCrypto at a glance | | SerpCrypto | SerpApi | | --- | --- | --- | | Billing model | Prepaid packs; credits never expire | Monthly plans; searches reset at the start of each billing cycle | | Smallest paid option | 10,000 searches for $7.50 | Starter: 1,000 searches a month for $25 | | 100,000 searches | $56.25 | $725 a month (Searcher) | | 1,000,000 searches | $375 | $3,750 a month (Cloud 1M) | | Free tier | None | 250 searches a month | | Failed searches | Not charged | Not counted ("Cached, errored, and failed searches are not") | | Engines | 6: Google Search, Google News, Google Images, Google Maps (local results), Google Shopping and Bing | Many more, including Google AI Mode | | Legal coverage | None offered | U.S. Legal Shield, up to $2 million, from the Production plan up | | Guaranteed throughput | Concurrency limit shown in the dashboard | Searches per hour per plan, e.g. 20,000 on Searcher | | Payment | BTC, USDT, USDC, ETH, SOL, LTC | No payment method listed on the pricing page | | Identity checks | No KYC: no ID, no selfie, no documents | — | *Prices checked on October 22, 2026.* SerpApi facts from serpapi.com/pricing, read on 2026-10-22. "—" = not stated on the pages we read. ## How much cheaper is SerpCrypto for the same number of searches? For the volumes SerpApi sells, our list price per search is lower at every tier we compared on 22 October 2026. The gap is widest on small volumes, where SerpApi's monthly plans cost $10 to $25 per 1,000 searches. | Searches | SerpApi plan (per month) | SerpCrypto pack | SerpCrypto per 1,000 | |---|---|---|---| | 5,000 | Developer, $75 | 10,000 searches, $7.50 | $0.75/1K searches | | 15,000 | Production, $150 | 25,000 searches, $18.75 | $0.75/1K searches | | 100,000 | Searcher, $725 | 100,000 searches, $56.25 | $0.56/1K searches | | 250,000 | Volume, $1,475 | 250,000 searches, $140.63 | $0.56/1K searches | | 500,000 | Infrastructure, $2,750 | 500,000 searches, $281.25 | $0.56/1K searches | | 1,000,000 | Cloud 1M, $3,750 | 1,000,000 searches, $375 | $0.38/1K searches | | 5,000,000 | Cloud 5M, $11,475 | 5,000,000 searches, $1,500 | $0.30/1K searches | The comparison is not like for like on two points. A SerpApi plan renews every month, while a SerpCrypto pack is bought once and lasts until it is used. And SerpApi's price includes a free tier, more engines and legal coverage on larger plans, which we do not offer. ## When is SerpApi the better choice? Choose SerpApi if one of these matters more than price or payment method: - **You need an engine we do not cover.** SerpApi lists many more engines and verticals than our 6, including Google AI Mode. - **You want legal coverage.** From its Production plan up, SerpApi includes a U.S. Legal Shield of up to $2 million "for the scraping and parsing of search engine data". SerpCrypto offers nothing comparable. - **You need a guaranteed hourly volume.** SerpApi publishes a throughput per plan, for example 20,000 successful searches an hour on Searcher. - **You want to start free.** SerpApi gives 250 searches a month without paying; we have no free tier. It is also the provider most often cited in comparisons, so libraries and tutorials tend to use its field names first. ## When is SerpCrypto the better choice? Choose SerpCrypto if you pay in crypto, prefer not to upload identity documents, or have uneven volumes. - **Crypto and no KYC.** Top up from $25 in BTC, USDT, USDC, ETH, SOL, LTC. Account-number login; email is optional. - **Credits that do not reset.** A monthly plan wastes what you do not use; a prepaid pack keeps it. That suits AI agents, research projects and seasonal SEO work. - **Price per search.** Our list prices are below SerpApi's at each plan size compared above, as of 22 October 2026. - **One price for every engine.** Google, Maps, News, Images, Shopping and Bing each cost one credit. ## How hard is it to switch from SerpApi? Switching is mostly a change of URL, key and a few names. Our response uses familiar block names such as `organic_results`, `people_also_ask`, `local_results`, `knowledge_graph` and `related_searches`, and requests take `q`, `gl`, `hl`, `location` and `device`. Check three things before you move a production job. First, the `engine` values we accept: `google`, `google_news`, `google_images`, `google_maps`, `google_shopping` and `bing`. Second, paging: we use `page=1` to `10` rather than an offset. Third, the fields your code reads: the [response format](https://serpcrypto.com/docs/response/) lists every field we return, and anything not listed should be treated as absent. Run your 50 most important queries on both services for a day and compare before switching the rest. *Python: the same query on SerpCrypto* ```python import os import requests params = { "engine": "google", "q": "best crm for small business", "gl": "us", "hl": "en", "location": "Chicago, Illinois, United States", } resp = requests.get( "https://serpcrypto.com/api/v1/search", headers={"Authorization": "Bearer " + os.environ["SERPCRYPTO_API_KEY"]}, params=params, timeout=60, ) resp.raise_for_status() data = resp.json() print([r["link"] for r in data.get("organic_results", [])][:5]) print([q["question"] for q in data.get("people_also_ask", [])]) ``` ## What about the lawsuits against SerpApi? SerpApi is a defendant in two US cases about SERP scraping, and both were open on 22 October 2026. In *Google LLC v. SerpApi, LLC* (N.D. Cal.), the court granted SerpApi's motion to dismiss on 20 July 2026, Google filed an amended complaint on 10 August, and no ruling on SerpApi's second motion to dismiss had been published. In *Reddit v. SerpApi, Oxylabs, AWMProxy and Perplexity* (S.D.N.Y.), the court largely denied SerpApi's and Perplexity's motions to dismiss on 31 July 2026. These are allegations, and no defendant has been found liable. They concern the whole SERP API industry, not one company: every provider, including SerpCrypto, collects public search results. Our [overview of the legal situation](https://serpcrypto.com/guides/is-a-serp-api-legal/) gives the timeline; it is not legal advice. ## Which pack replaces my SerpApi plan? Take your monthly searches on SerpApi. Up to 5,000 a month, 10,000 searches lasts two months or more; at 100,000 a month, 100,000 searches covers one month for $56.25. Credits never expire. | Plan | Quantity | Price | Price per unit | | --- | --- | --- | --- | | Test | 10,000 searches | $7.50 | $0.75/1K searches | | Build | 100,000 searches | $56.25 | $0.56/1K searches | | Volume | 1,000,000 searches | $375 | $0.38/1K searches | | Bulk | 5,000,000 searches | $1,500 | $0.30/1K searches | In every plan: Every engine at 1 credit per search · Failed requests not charged · Credits never expire · Country, language, city and device targeting **Before you pay** - Minimum top-up $25. Top up once in crypto, then buy any plan from your balance. - Unused balance stays on your account and never expires, and it pays for search credits. When a purchase costs less than the minimum top-up, the part added only to reach the minimum is refundable on request to support within 30 days of the payment, in USDT or USDC, as long as it has not been spent. The rest of the balance is not paid back, except where the law requires it. - Credited once your payment is checked on the blockchain. Typical network confirmation time: BTC ≈ 20 min · USDT TRC-20 ≈ 1 min · USDT ERC-20 ≈ 5 min · USDC ERC-20 ≈ 5 min · ETH ≈ 5 min · SOL ≈ 1 min · LTC ≈ 15 min ## SerpApi alternative: common questions ### Is SerpCrypto cheaper than SerpApi? At the list prices we read on 22 October 2026, yes, at every SerpApi plan size we compared: for example $725 a month for 100,000 searches on SerpApi Searcher, against $56.25 for 100,000 searches on SerpCrypto. SerpApi includes a free tier, more engines and legal coverage that we do not offer. ### Does SerpCrypto have a free plan like SerpApi? No. SerpApi gives 250 free searches a month. SerpCrypto has no free tier; the smallest pack is 10,000 searches for $7.50, paid from a $25 top-up. ### Can I pay SerpApi in crypto? SerpApi's pricing page, read on 22 October 2026, lists no payment method. SerpCrypto takes BTC, USDT, USDC, ETH, SOL, LTC only, with no KYC. ### Do unused SerpApi searches roll over? SerpApi's pricing page says searches reset at the start of each new billing cycle; on a downgrade, remaining searches move to an "Extra Credits" balance. SerpCrypto credits never expire. ### Does SerpCrypto support Google AI Mode? No. We return the Google AI Overview when the results page shows one, but we have no AI Mode engine. SerpApi lists one. ## Sources 1. [SerpApi pricing](https://serpapi.com/pricing) — SerpApi (read 2026-10-22) 2. [Google AI Mode API](https://serpapi.com/google-ai-mode-api) — SerpApi (read 2026-10-22) 3. [Court dismisses Google's DMCA claims against SerpApi](https://www.searchenginejournal.com/court-dismisses-googles-dmca-claims-against-serpapi/583033/) — Search Engine Journal (2026-07-21) 4. [SerpApi asks judge to end Google's scraping case](https://ppc.land/serpapi-asks-judge-to-end-googles-scraping-case-over-three-unseen-contracts/) — PPC Land (2026-08-25) 5. [Is scraping Google legal?](https://parallel.ai/articles/is-scraping-google-legal) — Parallel (2026-09-26) --- SerpCrypto: SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. Prepaid search credits from $0.30/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto. --- # Serper alternative: credits that never expire, paid in crypto > Serper alternative for 2026: Google and Bing results as JSON, credits that never expire instead of 6 months, no KYC, paid in crypto. Dated price table. - URL: https://serpcrypto.com/alternatives/serper/ - Last updated: 2026-10-22 - Publisher: SerpCrypto (https://serpcrypto.com) Serper is the usual reference for low-cost Google results, with a free start and fast responses. This page sets its four packs next to ours, shows where each one is the better buy, and explains what changes if you move. **Quick answer — What is a good Serper alternative that I can pay for in crypto?** **SerpCrypto** is a Serper alternative that returns Google and Bing results pages as JSON, paid in crypto with no KYC. Its credits never expire, where Serper's prepaid credits are valid for 6 months. On 22 October 2026, Serper sold 50,000 queries for $50 and 12.5 million for $3,750; SerpCrypto sells 100,000 searches for $56.25, down to $0.30/1K searches. ## Serper vs SerpCrypto at a glance | | SerpCrypto | Serper | | --- | --- | --- | | Billing model | Prepaid packs from a crypto balance | Prepaid packs | | Credit validity | Never expire | 6 months | | Smallest paid pack | 10,000 searches for $7.50 | 50,000 queries for $50 | | Largest self-serve pack | 5,000,000 searches for $1,500 | 12,500,000 queries for $3,750 | | Free start | None | 2,500 free queries, no credit card | | What is billed | 1 credit per successful search; failed requests free | Credits deducted for successful responses | | Engines | Google Search, News, Images, Maps, Shopping, and Bing | Google only: Search, Images, News, Maps, Places, Videos, Shopping, Scholar, Patents, Autocomplete | | Speed | Not published until measured | "Google search results in 1-2 seconds"; about 0.6–0.7 s median in Proxyway's January 2026 tests | | Rate limit | Concurrency limit shown in the dashboard | 50 to 300 queries per second, by pack | | Payment | BTC, USDT, USDC, ETH, SOL, LTC | Visa, Mastercard, American Express, Discover, PayPal | | Identity checks | No KYC: no ID, no selfie, no documents | None stated in the terms | | Refund | Within 72 hours if less than 20% used, to your balance | Within 7 days of first payment if less than 20% used | *Prices checked on October 22, 2026.* Serper facts from serper.dev and serper.dev/terms, read on 2026-10-22; speed figure from Proxyway's 2026 report. SerpCrypto prices are the ones on this page, set on 2026-10-22. ## How do the prices compare at each volume? At the prices on this page, set on 2026-10-22, our rate per 1,000 is below Serper's at every pack size up to 2.5 million, and the two meet at $0.30 per 1,000 on the largest packs. Serper's per-1,000 rate depends on the pack: $1.00, $0.75, $0.50 and $0.30. Its credits last 6 months, so a pack only pays off if you use it within that time. | Searches you need | Serper pack and rate | SerpCrypto pack and rate | |---|---|---| | 10,000 | Starter, $50 for 50,000 ($1.00 per 1,000) | 10,000 searches, $7.50 ($0.75/1K searches) | | 50,000 | Starter, $50 ($1.00 per 1,000) | 100,000 searches, $56.25 ($0.56/1K searches) covers it twice | | 500,000 | Standard, $375 ($0.75 per 1,000) | 500,000 searches, $281.25 ($0.56/1K searches) | | 2,500,000 | Scale, $1,250 ($0.50 per 1,000) | 1,000,000 searches ×2 and 500,000 searches, or 5,000,000 searches ($0.30/1K searches) | | 12,500,000 | Ultimate, $3,750 ($0.30 per 1,000) | 5,000,000 searches ×2.5 at $0.30/1K searches | At 12.5 million queries, Serper's Ultimate pack works out at $0.30 per 1,000, the same rate as our 5,000,000 searches pack, in one larger purchase. Below that, compare the rates line by line, and count how many of the credits you will really use within 6 months. ## When is Serper the better choice? Serper is the better buy in four situations: - **You want to try before paying.** Serper gives 2,500 free queries and asks for no credit card. SerpCrypto has no free tier. - **You need Google verticals we do not cover.** Serper lists Places, Videos, Scholar, Patents and Autocomplete endpoints; we cover Search, News, Images, Maps and Shopping. - **Speed is your first criterion.** Proxyway's report of 4 March 2026, based on tests in January, put Serper among the fastest real-time search APIs, with a median of about 0.6 to 0.7 seconds. We publish no speed figure until we have measured one. - **You buy more than 5 million searches at once.** Its $3,750 Ultimate pack holds 12.5 million searches at $0.30 per 1,000, the same rate as our largest pack, in a single purchase. ## When is SerpCrypto the better choice? SerpCrypto fits better when the payment method, the credit validity or Bing matters to you. - **You pay in crypto.** Top up from $25 in BTC, USDT, USDC, ETH, SOL, LTC. Serper lists cards and PayPal. - **You prefer not to share identity documents or a card.** Account-number login; email is optional. No ID, no selfie, no documents. - **Your usage is irregular.** A pack bought for a research project or an agent that runs in bursts keeps its value: Prepaid search credits never expire: unused searches stay on your account until you use them. - **You need Bing results too.** `engine=bing` returns `organic_results` with the same key and the same credit price. Serper covers Google only. - **You want the AI Overview with the web results.** Our Google response holds `ai_overview` with its sources when the page shows one. ## What changes in my code if I move from Serper? Three things change: the endpoint, the way you send the key and the names of a few response keys. SerpCrypto takes a `GET` request with query parameters such as `q`, `gl`, `hl` and `location`, and the key in an `Authorization: Bearer` header, or `api_key` in the query string. On the response side, read results from `organic_results`, questions from `people_also_ask` and the map pack from `local_results`. Each result carries a `position`, a `title`, a `link` and a `snippet`. The [response format](https://serpcrypto.com/docs/response/) lists every key, and [the search endpoint](https://serpcrypto.com/docs/search/) lists every parameter. Before you switch a production job, run your 50 most important queries on both services for a day and compare the outputs. *Python: one Google search on SerpCrypto* ```python import os import requests resp = requests.get( "https://serpcrypto.com/api/v1/search", headers={"Authorization": "Bearer " + os.environ["SERPCRYPTO_API_KEY"]}, params={"engine": "google", "q": "standing desk", "gl": "us", "hl": "en"}, timeout=60, ) resp.raise_for_status() results = resp.json() for item in results.get("organic_results", []): print(item["position"], item["title"], item["link"]) ``` ## Which pack replaces my Serper pack? Pick the pack closest to what you really use in a few months, not what you might use: with no expiry date, a smaller pack bought twice costs only a little more per search than a large one. | Plan | Quantity | Price | Price per unit | | --- | --- | --- | --- | | Test | 10,000 searches | $7.50 | $0.75/1K searches | | Build | 100,000 searches | $56.25 | $0.56/1K searches | | Scale | 500,000 searches | $281.25 | $0.56/1K searches | | Volume | 1,000,000 searches | $375 | $0.38/1K searches | In every plan: Every engine at 1 credit per search · Failed requests not charged · Credits never expire · Country, language, city and device targeting **Before you pay** - Minimum top-up $25. Top up once in crypto, then buy any plan from your balance. - Unused balance stays on your account and never expires, and it pays for search credits. When a purchase costs less than the minimum top-up, the part added only to reach the minimum is refundable on request to support within 30 days of the payment, in USDT or USDC, as long as it has not been spent. The rest of the balance is not paid back, except where the law requires it. - Credited once your payment is checked on the blockchain. Typical network confirmation time: BTC ≈ 20 min · USDT TRC-20 ≈ 1 min · USDT ERC-20 ≈ 5 min · USDC ERC-20 ≈ 5 min · ETH ≈ 5 min · SOL ≈ 1 min · LTC ≈ 15 min ## Serper alternative: common questions ### Do Serper credits expire? Yes. Serper's homepage, read on 22 October 2026, says its prepaid credits are valid for 6 months. SerpCrypto credits never expire: unused searches stay on your account until you use them. ### Can I pay Serper with crypto? Serper's homepage lists Visa, Mastercard, American Express, Discover and PayPal, and no crypto option. SerpCrypto accepts BTC, USDT, USDC, ETH, SOL, LTC only, from a $25 top-up, with no KYC. ### Is SerpCrypto cheaper than Serper? At the prices on this page, set on 2026-10-22, yes up to 2.5 million searches: for example $0.56/1K searches on our 100,000 searches pack against $0.75 on Serper's 500,000 pack. At the top, Serper's 12.5 million pack and our 5,000,000 searches pack both cost $0.30 per 1,000. Our credits never expire; Serper's last 6 months. ### Does SerpCrypto have Bing results like Serper? Serper covers Google only. SerpCrypto returns Bing results with `engine=bing`, for the same one credit per search as Google. ### Does SerpCrypto have a free tier like Serper? No. Serper offers 2,500 free queries without a card. No free tier: test with the smallest pack, paid from the minimum top-up; the part added only to reach the minimum is refundable on request within 30 days, in USDT or USDC, if unspent. ## Sources 1. [Serper: Google Search API](https://serper.dev/) — Serper (read 2026-10-22) 2. [Serper terms of service](https://serper.dev/terms) — Serper (read 2026-10-22) 3. [Search APIs 2026 report](https://proxyway.com/research/search-apis-2026-report) — Proxyway (2026-03-04, read 2026-10-22) --- SerpCrypto: SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. Prepaid search credits from $0.30/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto. --- # Tavily alternative: Google's results page as JSON for your agents > Tavily alternative for agents that need Google's own results: the results page as JSON from $0.30/1K searches, no KYC, crypto. Prices dated 2026-10-22. - URL: https://serpcrypto.com/alternatives/tavily/ - Last updated: 2026-10-22 - Publisher: SerpCrypto (https://serpcrypto.com) Tavily is an AI search API built for agents. Sometimes an agent needs something else: what Google actually shows for a query, in a given country, with its AI Overview sources and map pack. This page explains the difference and when to use each. **Quick answer — What is a good Tavily alternative when my agent needs real Google results?** Use a SERP API. **SerpCrypto** returns the Google or Bing results page as JSON, with organic results, People Also Ask and the AI Overview sources, for one credit per search: 100,000 searches cost $56.25 ($0.56/1K searches). Tavily, an AI search API, charged $7.50 per 1,000 basic searches on its Project plan on 22 October 2026. SerpCrypto needs no KYC and takes crypto. ## Tavily on 22 October 2026 | Attribute | Value | | --- | --- | | What it is | An AI search API for agents, branded "Tavily by Nebius" | | Plans | Free 1,000 credits a month; Project $30 for 4,000; Bootstrap $100 for 15,000; Startup $220 for 38,000; Growth $500 for 100,000 | | Pay as you go | $0.008 per credit | | Credits per search | Basic search 1 credit, advanced search 2 | | Price per 1,000 basic searches | $7.50 on Project, $5.00 on Growth, $8.00 pay as you go | | Credit reset | "Your API credits reset on the first day of each month" | | Payment | Credit cards, debit cards, bank transfers; x402 at $0.01 per call in USDC on Base, no account | | Ownership | Nebius announced an agreement to acquire Tavily on 10 February 2026; the value was not disclosed | ## What is the difference between Tavily and a SERP API? Tavily runs its own retrieval and returns results prepared for a language model. A SERP API returns the results page of a public search engine, exactly as that engine laid it out for a query, a place and a device. The two answer different questions. An AI search API answers "what does the web say about this?" in a form a model can read. A SERP API answers "what does Google show for this?": which pages rank, in what order, which sources the AI Overview cites, which businesses sit in the map pack. If your agent reports rankings, checks a brand's visibility or cites what a person would see on Google, it needs the second answer. ## How do AI search APIs and a SERP API compare on price? List prices per 1,000 searches, read on each provider's own page. AI search APIs use their own index or retrieval; SerpCrypto returns Google and Bing pages. | Provider | What it returns | Price per 1,000 | Crypto | | --- | --- | --- | --- | | SerpCrypto | Google and Bing results pages as JSON | $0.75/1K searches down to $0.30/1K searches | BTC, USDT, USDC, ETH, SOL, LTC | | Tavily | AI search results from its own retrieval | $7.50 (Project) to $5.00 (Growth); $8.00 pay as you go | x402 USDC, $0.01 per call | | Exa | Results from its own index | $4 lite, $7 auto | x402 USDC, $0.004–$0.015 per request | | Brave Search API | Results from Brave's own index | $5 | Not stated; a card is required | | You.com Web Search | Up to 100 results per call | $5 | x402 and MPP in USDC | | Perplexity Search API | Results from its own index | $5 ($1 in fast mode) | Not stated | | Linkup | Standard search | $5 | x402 USDC on Base | *Prices checked on October 22, 2026.* Sources at the bottom of the page. Credits, modes and extras differ: read each provider's page before deciding. ## When should I keep Tavily? Keep Tavily, or another AI search API, when the agent only needs relevant text and does not care where Google would rank it. - **Research agents** that read and summarise many pages benefit from retrieval tuned for models. - **Agents that pay per call without an account** can use Tavily's x402 option at $0.01 per call in USDC. - **Speed-sensitive chat** may prefer Tavily's own latency claim of 180 ms at the median; we publish no speed figure until we have measured one. The two kinds of tool can also work together: an AI search API for reading, and a SERP API for facts about Google itself. ## When do Google's own results matter for an agent? Google's page matters whenever the answer depends on Google, not on the web in general. 1. **Visibility questions:** "does our page rank for this?", "who ranks above us in Germany?" 2. **AI Overview sources:** which pages Google's overview cites, from `ai_overview.references`. 3. **Local answers:** the businesses in the map pack for a query in a given city, from `local_results`. 4. **What a person would see:** questions in People Also Ask, top stories, the knowledge panel. 5. **Rank data over time:** positions you can store and compare day after day. With SerpCrypto, each of these is one search and one credit, on Google or Bing. *Python: turn a results page into a compact context for a model* ```python import os import requests def google_context(query, gl="us", hl="en", max_results=5): """Search once and return a short text block a model can read.""" resp = requests.get( "https://serpcrypto.com/api/v1/search", headers={"Authorization": "Bearer " + os.environ["SERPCRYPTO_API_KEY"]}, params={"engine": "google", "q": query, "gl": gl, "hl": hl}, timeout=60, ) resp.raise_for_status() data = resp.json() lines = [] overview = data.get("ai_overview") if overview: cited = ", ".join(ref.get("link", "") for ref in overview.get("references", [])[:3]) lines.append("AI Overview cites: " + cited) for r in data.get("organic_results", [])[:max_results]: lines.append("[%d] %s | %s | %s" % (r["position"], r["title"], r["link"], r.get("snippet", ""))) for q in data.get("people_also_ask", [])[:3]: lines.append("People also ask: " + q["question"]) return "\n".join(lines) if __name__ == "__main__": print(google_context("best project management tool for agencies")) ``` ## What does the Nebius deal change for Tavily users? Nebius, an AI cloud company, announced an agreement to acquire Tavily in a press release dated 10 February 2026. The release says the transaction value was not disclosed and that Tavily will continue operating under its current brand. Tavily's pages now use the name "Tavily by Nebius". For a user, nothing in the public pricing pages we read on 22 October 2026 points to a change of plans or terms. If supplier concentration is a concern for your agent, keeping a second search source, such as a SERP API, is a simple hedge. ## How many credits does an agent need? Count one search per tool call. An agent that searches 300 times a day needs about 9,000 searches a month, which 10,000 searches covers. Credits never expire, so quiet weeks cost nothing. | Plan | Quantity | Price | Price per unit | | --- | --- | --- | --- | | Test | 10,000 searches | $7.50 | $0.75/1K searches | | Starter | 25,000 searches | $18.75 | $0.75/1K searches | | Build | 100,000 searches | $56.25 | $0.56/1K searches | | Volume | 1,000,000 searches | $375 | $0.38/1K searches | In every plan: Every engine at 1 credit per search · Failed requests not charged · Credits never expire · Country, language, city and device targeting **Before you pay** - Minimum top-up $25. Top up once in crypto, then buy any plan from your balance. - Unused balance stays on your account and never expires, and it pays for search credits. When a purchase costs less than the minimum top-up, the part added only to reach the minimum is refundable on request to support within 30 days of the payment, in USDT or USDC, as long as it has not been spent. The rest of the balance is not paid back, except where the law requires it. - Credited once your payment is checked on the blockchain. Typical network confirmation time: BTC ≈ 20 min · USDT TRC-20 ≈ 1 min · USDT ERC-20 ≈ 5 min · USDC ERC-20 ≈ 5 min · ETH ≈ 5 min · SOL ≈ 1 min · LTC ≈ 15 min ## Tavily alternative: common questions ### Is SerpCrypto a replacement for Tavily? For Google and Bing results, yes. For model-ready text from its own retrieval, no: that is what Tavily does. SerpCrypto returns the results page as JSON, and your code or model decides what to read next. ### Can AI agents pay for Tavily in crypto? Yes. Tavily's blog describes an x402 option at $0.01 per call in USDC on Base, with no account. SerpCrypto works differently: you top up a balance in BTC, USDT, USDC, ETH, SOL, LTC from $25 and buy credits that never expire. ### Do Tavily credits roll over? No. Tavily's terms say API credits reset on the first day of each month. SerpCrypto credits never expire. ### Can I use SerpCrypto with MCP like Tavily? Yes, through a small MCP server you run yourself. The [MCP page](https://serpcrypto.com/docs/mcp/) has the code for Claude Desktop, Cursor and other MCP clients, and the [LangChain page](https://serpcrypto.com/docs/langchain/) has a tool definition. ### Who owns Tavily? Nebius announced an agreement to acquire Tavily on 10 February 2026, with the value not disclosed. Tavily's pages now say "Tavily by Nebius". ## Sources 1. [Tavily API credits](https://docs.tavily.com/documentation/api-credits) — Tavily (read 2026-10-22) 2. [Tavily terms of service](https://www.tavily.com/terms) — Tavily (read 2026-10-22) 3. [Tavily on x402](https://tavily.com/blog/tavily-on-x402) — Tavily (read 2026-10-22) 4. [Nebius announces agreement to acquire Tavily](https://nebius.com/newsroom/nebius-announces-agreement-to-acquire-tavily-to-add-agentic-search-to-its-ai-cloud-platform) — Nebius (2026-02-10, read 2026-10-22) 5. [Exa pricing](https://exa.ai/pricing) — Exa (read 2026-10-22) 6. [Exa x402 guide](https://exa.ai/docs/reference/x402-guide) — Exa (read 2026-10-22) 7. [Brave Search API](https://brave.com/search/api/) — Brave (read 2026-10-22) 8. [You.com pricing](https://you.com/pricing) — You.com (read 2026-10-22) 9. [You.com machine payments](https://you.com/docs/administration/machine-payments) — You.com (read 2026-10-22) 10. [Perplexity Search API overview](https://docs.perplexity.ai/docs/search/overview.md) — Perplexity (read 2026-10-22) 11. [Linkup pricing](https://docs.linkup.so/pages/documentation/development/pricing) — Linkup (read 2026-10-22) 12. [Linkup x402](https://docs.linkup.so/pages/documentation/platform/x402) — Linkup (read 2026-10-22) --- SerpCrypto: SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. Prepaid search credits from $0.30/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto. --- # Best search APIs for AI agents in 2026: 9 options compared > Best web search APIs for AI agents and LLMs in 2026: Google SERP APIs vs Tavily, Exa, Brave, Perplexity. Price per 1,000, crypto, x402, KYC. Checked 2026-10-22. - URL: https://serpcrypto.com/best/search-apis-for-ai-agents/ - Last updated: 2026-10-22 - Publisher: SerpCrypto (https://serpcrypto.com) An agent's web search tool can return Google's results page or an answer from a provider's own index. This page compares nine APIs on price, output, free tier, payment (including per-call crypto) and identity checks. We publish SerpCrypto, which is listed first. **Quick answer — What is the best web search API for AI agents in 2026?** It depends on what the agent needs. For Google's own results, rankings and AI Overview sources, use a SERP API: **SerpCrypto** charges $0.75/1K searches down to $0.30/1K searches, paid in crypto with no KYC. For answers from a dedicated index, Brave, Perplexity, Linkup and You.com list $5.00 per 1,000 and Exa from $4.00, as of 22 October 2026; Tavily lists $7.50 at entry. ## The 9 best search APIs for AI agents in 2026 **Methodology.** Published by SerpCrypto (SerpGoat LLC). Every provider is compared on the same criteria, taken from its own public website: list price per 1,000 searches, what the API returns (a search engine's results page or its own index), free tier, accepted payment methods, including per-call crypto (x402), identity or card requirements, the date the prices were checked. *Prices checked on October 22, 2026.* ### 1. SerpCrypto — Our product Google and Bing results pages as JSON for agents that need what users actually see: rankings, People Also Ask, local results and AI Overview sources. - Best for: Agents grounded in Google results; builders who pay in crypto - Price: $0.75/1K searches on 10,000 searches; $0.30/1K searches on 5,000,000 searches - KYC: None - Crypto payment: BTC, USDT, USDC, ETH, SOL, LTC (prepaid balance, not per call) - Pros: Credits never expire - Pros: Ready tool code for MCP, LangChain, OpenAI, n8n and OpenClaw - Cons: No free tier - Cons: Returns results, not page content or written answers - Cons: No per-call x402 payment ### 2. Tavily A search API built for LLM agents, which returns content prepared for a model. Nebius agreed to acquire Tavily in February 2026; Tavily keeps its brand. - Best for: Agent frameworks that expect LLM-ready content - Price: $7.50 per 1,000 (Project, $30 for 4,000 credits); $5.00 on Growth ($500 for 100,000); $8.00 pay as you go - KYC: Not stated - Crypto payment: USDC per call via x402, $0.01 a call - Pros: 1,000 free credits a month - Pros: Per-call crypto without an account - Cons: Monthly credits reset - Cons: Not Google's results page ### 3. Exa A search engine with its own index, designed for AI applications, with two search modes priced separately. - Best for: Semantic search over its own web index - Price: $4.00 per 1,000 (lite) or $7.00 (auto) - KYC: Not stated - Crypto payment: Per-request USDC via x402 - Pros: $10 of free usage a month - Pros: Among the fastest indices in Proxyway's March 2026 test - Cons: Not Google's results page ### 4. Brave Search API Results from Brave's independent index at a flat price, with a monthly credit. - Best for: A non-Google index at a predictable flat price - Price: $5.00 per 1,000; $5 free credit a month (about 1,000 requests) - KYC: Credit card required, "to prevent abuse" - Crypto payment: Not stated - Pros: Own independent index - Pros: Flat price - Cons: Card required even for the free credit ### 5. Perplexity Search API Search results from Perplexity's index, with a cheaper fast mode. - Best for: Agents that want Perplexity's retrieval - Price: $5.00 per 1,000; $1.00 in fast mode - KYC: Not stated - Crypto payment: Not stated - Pros: Among the fastest indices in Proxyway's March 2026 test - Pros: Low-cost fast mode - Cons: Not Google's results page ### 6. Linkup A search API for AI with standard and deep modes, payable per request in USDC without an account. - Best for: Agents that pay per request with no signup - Price: $5.00 per 1,000 (standard); $50.00 (deep) - KYC: Professional email for the free monthly credit; none via x402 - Crypto payment: USDC on Base via x402, $0.01 a request - Pros: No account needed with x402 - Pros: Monthly free credit - Cons: Deep mode costs 10 times standard ### 7. You.com Web Search API A web search API returning up to 100 results per call, with machine payments in stablecoins. - Best for: Many results per call; agents paying in stablecoins - Price: $5.00 per 1,000 calls - KYC: None for machine payments - Crypto payment: USDC via x402 (Base, Solana) and MPP - Pros: Up to 100 results per call - Pros: $10 free credit (about 2,000 searches) - Cons: Not Google's results page ### 8. Serper A low-cost Google SERP API, the usual low-price reference in published SERP API comparisons. - Best for: High-volume Google results for agents paid by card - Price: $1.00 per 1,000 down to $0.30 at 12.5 million - KYC: None stated - Crypto payment: Not mentioned - Pros: Low price at volume - Pros: 2,500 free queries, no card - Cons: Credits expire after 6 months ### 9. Firecrawl search Web search combined with page scraping, for agents that need the content of the pages they find. - Best for: Search plus full-page content in one call - Price: $7.60 per 1,000 at entry (2 credits per search) down to $1.50 on the Scale plan - KYC: Not stated - Crypto payment: Not stated - Pros: Returns page content - Pros: 1,000 free credits a month - Cons: Plan credits do not roll over *Disclosure: this page is published by SerpCrypto. Competitor details come from their public websites at the date shown above and may have changed since.* ## Does my agent need a SERP API or an AI search API? Ask what the agent has to know. If it must know what Google shows, choose a SERP API. That covers rankings, ads, local results, the AI Overview and its sources, and fresh news cards: things only the engine's own page holds. If it needs relevant text to answer a question, an AI search API is often simpler. Tavily, Exa, Brave, Perplexity, Linkup and You.com answer from their own index or retrieval, and some return text ready for the model. Many agents use both: a SERP API to find and rank sources as people see them, and a fetch or extract step to read the pages. The [AI agents guide](https://serpcrypto.com/use-cases/ai-agents/) shows how to keep the SERP part small in tokens. ## Can an AI agent pay for each search in crypto? Yes, with some AI search APIs. Linkup, You.com, Tavily and Exa accept per-request payments in USDC through x402, an HTTP payment protocol: the agent pays for a call without an account. Linkup lists $0.01 a request, Tavily $0.01 a call, and You.com $0.005 a call on x402. SerpCrypto works differently: you top up a balance once in BTC, USDT, USDC, ETH, SOL, LTC, buy credits that never expire, and the agent uses an API key. That suits agents run by a team; per-call x402 suits fully autonomous agents. We do not offer x402 today. ## Which search APIs are fastest for agents? The most-cited independent test is Proxyway's report of 4 March 2026, which ran 15 services three times with 2,000 requests. It found Perplexity, Exa and Tavily the fastest of the index-based APIs, and Oxylabs and Serper the fastest live SERP APIs, at a median of about 0.6 to 0.7 seconds. We have not published speed figures for SerpCrypto; we will add them with a method and a date once measured. For an agent, total latency also depends on how much text you pass to the model, so trimming the response matters as much as the API. ## What does SerpCrypto cost for an agent? An agent that runs 300 searches a day uses about 9,000 a month: 10,000 searches covers it for $7.50. Credits never expire, so idle weeks cost nothing. | Plan | Quantity | Price | Price per unit | | --- | --- | --- | --- | | Test | 10,000 searches | $7.50 | $0.75/1K searches | | Starter | 25,000 searches | $18.75 | $0.75/1K searches | | Build | 100,000 searches | $56.25 | $0.56/1K searches | | Volume | 1,000,000 searches | $375 | $0.38/1K searches | In every plan: Every engine at 1 credit per search · Failed requests not charged · Credits never expire · Country, language, city and device targeting **Before you pay** - Minimum top-up $25. Top up once in crypto, then buy any plan from your balance. - Unused balance stays on your account and never expires, and it pays for search credits. When a purchase costs less than the minimum top-up, the part added only to reach the minimum is refundable on request to support within 30 days of the payment, in USDT or USDC, as long as it has not been spent. The rest of the balance is not paid back, except where the law requires it. - Credited once your payment is checked on the blockchain. Typical network confirmation time: BTC ≈ 20 min · USDT TRC-20 ≈ 1 min · USDT ERC-20 ≈ 5 min · USDC ERC-20 ≈ 5 min · ETH ≈ 5 min · SOL ≈ 1 min · LTC ≈ 15 min ## Search APIs for AI agents: common questions ### Which search API do Claude, ChatGPT and Cursor agents use? Any API wrapped as a tool. Through MCP, Claude Desktop, Cursor and other clients can call a search tool you configure. Our [MCP page](https://serpcrypto.com/docs/mcp/) has a small server, and [OpenClaw](https://serpcrypto.com/docs/openclaw/) and [n8n](https://serpcrypto.com/docs/n8n/) have their own guides. ### Is there a free web search API for AI agents? Several have free allowances: Tavily 1,000 credits a month, Exa $10 a month, Brave a $5 monthly credit with a card on file, and Serper 2,500 queries once. SerpCrypto has no free tier; its smallest pack is 10,000 searches for $7.50. ### Which search API works without a credit card? Serper's free queries need no card, and Linkup, You.com, Tavily and Exa accept per-call USDC via x402. SerpCrypto never asks for a card: you pay in BTC, USDT, USDC, ETH, SOL, LTC with no KYC. ### What is x402? An HTTP-based protocol for paying per request in stablecoins such as USDC, without an account. Several AI search APIs accept it for agent payments. SerpCrypto does not offer it today. ### How many tokens does a search result cost an agent? It depends on what you pass on. Sending only the title, link and snippet of 10 organic results keeps the tool output short. The [AI agents guide](https://serpcrypto.com/use-cases/ai-agents/) shows a helper that trims our JSON before it reaches the model. ## Sources 1. [Search APIs 2026 report](https://proxyway.com/research/search-apis-2026-report) — Proxyway (2026-03-04) 2. [API credits](https://docs.tavily.com/documentation/api-credits) — Tavily (read 2026-10-22) 3. [Tavily on x402](https://tavily.com/blog/tavily-on-x402) — Tavily (read 2026-10-22) 4. [Nebius announces agreement to acquire Tavily](https://nebius.com/newsroom/nebius-announces-agreement-to-acquire-tavily-to-add-agentic-search-to-its-ai-cloud-platform) — Nebius (2026-02-10) 5. [Pricing](https://exa.ai/pricing) — Exa (read 2026-10-22) 6. [x402 guide](https://exa.ai/docs/reference/x402-guide) — Exa (read 2026-10-22) 7. [Brave Search API](https://brave.com/search/api/) — Brave (read 2026-10-22) 8. [Search API overview](https://docs.perplexity.ai/docs/search/overview.md) — Perplexity (read 2026-10-22) 9. [Pricing](https://docs.linkup.so/pages/documentation/development/pricing) — Linkup (read 2026-10-22) 10. [x402](https://docs.linkup.so/pages/documentation/platform/x402) — Linkup (read 2026-10-22) 11. [Pricing](https://you.com/pricing) — You.com (read 2026-10-22) 12. [Machine payments](https://you.com/docs/administration/machine-payments) — You.com (read 2026-10-22) 13. [Serper](https://serper.dev/) — Serper (read 2026-10-22) 14. [Pricing](https://www.firecrawl.dev/pricing) — Firecrawl (read 2026-10-22) --- SerpCrypto: SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. Prepaid search credits from $0.30/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto. --- # Best SERP APIs in 2026: 10 Google results APIs compared > Best SERP APIs in 2026: 10 Google SERP APIs compared on price per 1,000 searches, free tier, credit expiry, crypto and KYC. Prices checked on 2026-10-22. - URL: https://serpcrypto.com/best/serp-apis/ - Last updated: 2026-10-22 - Publisher: SerpCrypto (https://serpcrypto.com) Ten SERP APIs that return Google results as JSON, compared on the same criteria from their own public pages. We publish SerpCrypto, which is first on this list; each entry says what the provider does better than the others. **Quick answer — What is the best SERP API in 2026?** There is no single best SERP API; the right one depends on volume, engines and how you pay. On 22 October 2026: **SerpCrypto** for crypto payment, no KYC and credits that never expire ($0.75/1K searches down to $0.30/1K searches); Serper for low-cost Google results at large volume ($1.00 to $0.30 per 1,000); SerpApi for the widest engine list and legal coverage; DataForSEO for delayed bulk data at $0.60 per 1,000. ## The 10 best SERP APIs in 2026 **Methodology.** Published by SerpCrypto (SerpGoat LLC). Every provider is compared on the same criteria, taken from its own public website: list price per 1,000 Google searches at entry and at high volume, free tier, credit expiry, accepted payment methods, including crypto, identity checks (KYC), engines covered, the date the prices were checked. *Prices checked on October 22, 2026.* ### 1. SerpCrypto — Our product A SERP API for developers and AI agents, sold as prepaid credits that never expire, paid in crypto with no KYC. One credit per successful search on 6 engines. - Best for: Paying in crypto without uploading documents; uneven or seasonal volumes - Price: $0.75/1K searches on 10,000 searches, $0.38/1K searches on 1,000,000 searches, $0.30/1K searches on 5,000,000 searches - KYC: None - Crypto payment: BTC, USDT, USDC, ETH, SOL, LTC - Pros: Credits never expire - Pros: Same price on every engine - Pros: Tool definitions for MCP, LangChain and OpenAI - Cons: No free tier - Cons: 6 engines; no Google AI Mode - Cons: No legal coverage offered ### 2. Serper A Google-focused SERP API with low list prices that fall with volume, and 2,500 free queries without a card. - Best for: Large Google volumes at a low price per search - Price: $1.00 per 1,000 ($50 for 50,000) down to $0.30 ($3,750 for 12.5 million) - KYC: None stated - Crypto payment: Not mentioned - Pros: Low price at volume - Pros: 2,500 free queries, no card - Pros: Among the fastest in Proxyway's March 2026 test - Cons: Credits expire 6 months after purchase - Cons: Cards and PayPal only ### 3. SerpApi The best-known SERP API, with a long list of engines including Google AI Mode, hourly throughput guarantees and a U.S. Legal Shield on larger plans. - Best for: Many engines in one API; teams that want legal coverage - Price: $25 per 1,000 (Starter) to $3.75 at 1 million a month; $1.96 at 54 million - KYC: None stated - Crypto payment: Not mentioned - Pros: Widest engine coverage - Pros: U.S. Legal Shield up to $2 million from Production - Pros: 250 free searches a month - Cons: Monthly plans reset unused searches - Cons: Highest price per search at small volumes ### 4. DataForSEO Pay-as-you-go SERP data for SEO tools, with a low price for queued results and funds that never expire. - Best for: Bulk SEO data where a short delay is acceptable - Price: $0.60 per 1,000 (standard queue), $1.20 (priority), $2.00 (live); $50 minimum payment - KYC: None stated - Crypto payment: Bitcoin on request via support - Pros: Low queued price - Pros: Funds never expire - Pros: Billing per 10 results documented since September 2025 - Cons: Live results cost $2.00 per 1,000 - Cons: $50 minimum payment ### 5. Scrapingdog A scraping API with a Google endpoint whose price per search falls steeply on large plans. - Best for: Very large monthly volumes - Price: $1.00 per 1,000 on Lite ($40) with the 5-credit Light search; $0.35 at 1 million a month - KYC: None stated - Crypto payment: Not mentioned - Pros: Low price at 1 million a month and above - Pros: Credits roll over from the Standard plan up - Cons: 5 or 10 credits per Google search, depending on the request type - Cons: Lite plan credits end with the cycle ### 6. Decodo Web Data API A general web data API from a proxy provider, with a `/search` endpoint for Google and self-serve crypto payment. - Best for: Teams already on Decodo, or who want crypto from a large provider - Price: $0.75 per 1,000 on the $19 plan down to $0.40 on the $1,499 plan - KYC: Email verification - Crypto payment: Yes, self-serve (coins not stated) - Pros: Crypto accepted - Pros: Low price on large plans - Cons: Monthly plans - Cons: Coins and processor not named on the pricing page ### 7. Bright Data SERP API An enterprise-grade SERP API with pay-as-you-go pricing and a recurring free tier. - Best for: Enterprises with procurement and compliance teams - Price: $1.50 per 1,000 pay as you go; $1.31 on the $499 Scale plan - KYC: Not stated for the SERP API - Crypto payment: No - Pros: 5,000 free requests every month - Pros: Pay as you go - Cons: States that it does not accept crypto ### 8. Oxylabs Web Scraper API A large provider's scraping API with Google support and some of the fastest median response times in an independent 2026 test. - Best for: Low latency at scale, with a vendor that verifies customers - Price: $1.00 per 1,000 without JavaScript rendering on Starter ($30), $0.70 on Flexible ($500) - KYC: KYC questionnaire for every customer - Crypto payment: No - Pros: Among the fastest in Proxyway's March 2026 test - Pros: Clear plan ladder - Cons: KYC for everyone - Cons: No crypto - Cons: $1.35 per 1,000 with JavaScript rendering on Starter ### 9. SearchApi A SERP API with many engines and a dedicated rank tracking API since Google stopped serving 100 results per page. - Best for: Teams that want SerpApi-style coverage at a lower price - Price: $4.00 per 1,000 ($40 for 10,000) down to $1.00 ($5,000 for 5 million) - KYC: None stated - Crypto payment: Not mentioned - Pros: Wide engine list - Pros: Legal-liability coverage advertised on mid and upper plans - Cons: Higher entry price than Google-only APIs ### 10. Scrapeless A scraping platform with a pay-as-you-go Google endpoint and crypto payment. - Best for: Small pay-as-you-go volumes paid in crypto - Price: $1.00 per 1,000 pay as you go; $0.80 on the $399 Business plan - KYC: None for basic packages - Crypto payment: Yes (coins not stated) - Pros: Crypto accepted - Pros: No KYC on basic packages - Cons: Unused monthly minimum is not kept - Cons: Coins not named on the pricing page *Disclosure: this page is published by SerpCrypto. Competitor details come from their public websites at the date shown above and may have changed since.* ## How do I choose a SERP API? Start from the job, then compare price. Five questions settle most choices: 1. **How many searches a month, and how steady?** Steady, large volumes favour monthly plans with low unit prices. Uneven volumes favour credits that do not expire. 2. **Do you need results now or in a few minutes?** Queued results cost less: DataForSEO lists $0.60 per 1,000 queued and $2.00 live. 3. **Which engines?** Google-only APIs are cheaper. Broad catalogues such as SerpApi and SearchApi cover many more engines. 4. **How will you pay?** Most providers take cards only. Decodo, Scrapeless and SerpCrypto state crypto payment; DataForSEO takes Bitcoin on request. 5. **What checks will you accept?** Oxylabs runs a KYC questionnaire for every customer; SerpCrypto asks for no ID at all. Whatever you pick, test your 50 most important queries for a day before moving production traffic. ## Which SERP APIs are fastest? We do not publish our own speed figures until we have measured them on a public sample. The most-cited independent test is Proxyway's search API report of 4 March 2026: 15 services, three rounds of 2,000 requests in mid-January 2026. It found Oxylabs and Serper the fastest live SERP APIs, at a median of about 0.6 to 0.7 seconds. Vendor-run tests disagree with each other, and most rank their own API first. Treat any speed claim, including future ones from us, as a starting point for your own test. ## What changed for SERP APIs in 2025 and 2026? Three changes affect every provider on this list. Google stopped honouring `num=100` around 10 to 14 September 2025, so 100 results now take 10 searches. Google started routing result links through a `/goto` redirect, which it confirmed on 26 August 2026. And since about 11 September 2026, rank-tracking vendors have reported wide blocks and altered results pages; Nozzle measured 8.0% of pages affected. The official alternatives are closing too: Microsoft retired the Bing Search APIs on 11 August 2025, and Google's Custom Search JSON API ends on 1 January 2027. Our [guide to Google's num=100 change](https://serpcrypto.com/guides/google-100-results-change/) and [legal overview](https://serpcrypto.com/guides/is-a-serp-api-legal/) give the dated details. ## What does SerpCrypto cost? Prepaid credits, one per successful search on any of the 6 engines, never expiring. Prices set on 22 October 2026. | Plan | Quantity | Price | Price per unit | | --- | --- | --- | --- | | Test | 10,000 searches | $7.50 | $0.75/1K searches | | Build | 100,000 searches | $56.25 | $0.56/1K searches | | Volume | 1,000,000 searches | $375 | $0.38/1K searches | | Bulk | 5,000,000 searches | $1,500 | $0.30/1K searches | In every plan: Every engine at 1 credit per search · Failed requests not charged · Credits never expire · Country, language, city and device targeting **Before you pay** - Minimum top-up $25. Top up once in crypto, then buy any plan from your balance. - Unused balance stays on your account and never expires, and it pays for search credits. When a purchase costs less than the minimum top-up, the part added only to reach the minimum is refundable on request to support within 30 days of the payment, in USDT or USDC, as long as it has not been spent. The rest of the balance is not paid back, except where the law requires it. - Credited once your payment is checked on the blockchain. Typical network confirmation time: BTC ≈ 20 min · USDT TRC-20 ≈ 1 min · USDT ERC-20 ≈ 5 min · USDC ERC-20 ≈ 5 min · ETH ≈ 5 min · SOL ≈ 1 min · LTC ≈ 15 min ## Best SERP API: common questions ### Which SERP API is the most affordable? It depends on volume. On 22 October 2026, DataForSEO's standard queue listed $0.60 per 1,000 for delayed results; for live results, Serper and Scrapeless listed $1.00 at 10,000 a month and Scrapingdog $0.35 at 1 million. The [price index](https://serpcrypto.com/data/serp-api-price-index-2026/) has 19 providers. ### Which SERP API has a free tier? Bright Data gives 5,000 requests a month, SerpApi 250 a month and Serper 2,500 once, without a card. SerpCrypto has no free tier; its smallest pack is 10,000 searches for $7.50. ### Which SERP APIs accept crypto? Decodo and Scrapeless state self-serve crypto payment, DataForSEO takes Bitcoin on request, and SerpCrypto takes BTC, USDT, USDC, ETH, SOL, LTC. Bright Data and Oxylabs state that they do not accept crypto. ### Is this ranking independent? No. It is published by SerpCrypto, which is ranked first. Every provider is described from its own public pages, read on 22 October 2026, with the same criteria, and each entry lists what that provider does better. Proxyway's March 2026 report is a widely cited independent test. ### Which SERP API is best for AI agents? One that returns compact JSON and has ready tool definitions. See our separate ranking of [search APIs for AI agents](https://serpcrypto.com/best/search-apis-for-ai-agents/), which also covers AI search APIs such as Tavily, Exa and Brave. ## Sources 1. [Search APIs 2026 report](https://proxyway.com/research/search-apis-2026-report) — Proxyway (2026-03-04) 2. [SerpApi pricing](https://serpapi.com/pricing) — SerpApi (read 2026-10-22) 3. [Serper](https://serper.dev/) — Serper (read 2026-10-22) 4. [SERP API pricing](https://dataforseo.com/apis/serp-api/pricing) — DataForSEO (read 2026-10-22) 5. [Pricing](https://www.scrapingdog.com/pricing/) — Scrapingdog (read 2026-10-22) 6. [Pricing](https://decodo.com/pricing) — Decodo (read 2026-10-22) 7. [SERP API pricing](https://www.bright.cn/pricing/serp) — Bright Data (read 2026-10-22) 8. [Web Scraper API pricing](https://oxylabs.io/pricing/web-scraper-api) — Oxylabs (read 2026-10-22) 9. [KYC and safety](https://oxylabs.io/kyc-and-safety) — Oxylabs (read 2026-10-22) 10. [Pricing](https://www.searchapi.io/pricing) — SearchApi (read 2026-10-22) 11. [Pricing](https://www.scrapeless.com/en/pricing) — Scrapeless (read 2026-10-22) 12. [Google feeds rank trackers swapped results on 8% of pages, Nozzle finds](https://ppc.land/google-feeds-rank-trackers-swapped-results-on-8-of-pages-nozzle-finds/) — PPC Land (2026-10-19) --- SerpCrypto: SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. Prepaid search credits from $0.30/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto. --- # Bing Search API: Bing web results as JSON, after the 2025 retirement > Bing Search API after Microsoft's August 2025 retirement: Bing web results as JSON, 1 credit per search from $0.30/1K searches. No KYC, pay in crypto. - URL: https://serpcrypto.com/bing-search-api/ - Last updated: 2026-10-22 - Publisher: SerpCrypto (https://serpcrypto.com) Microsoft closed its Bing Search APIs in August 2025 and now sells grounded answers for AI agents instead. If you need the results page itself, set engine=bing and read it as JSON. **Quick answer — Is there still a Bing Search API that returns web results as JSON?** Microsoft retired its own Bing Search APIs on 11 August 2025. **SerpCrypto** offers a Bing engine in its SERP API: set `engine=bing` and you get Bing's web results page as JSON, with organic results and related searches, for one credit per successful search, from $0.30/1K searches as of October 2026. No KYC, no Azure account; pay in crypto. ## The SerpCrypto Bing engine at a glance | Attribute | Value | | --- | --- | | Endpoint | `GET https://serpcrypto.com/api/v1/search?engine=bing&q=…` | | Blocks returned | `organic_results` (position, title, link, displayed link, snippet) and `related_searches` | | Targeting | `gl` (country), `hl` (language), `location` (city), `device` (desktop or mobile) | | Pages | `page=1` to `page=10`, one credit each | | Price | $7.50 for 10,000 searches, $56.25 for 100,000 searches, from $0.30/1K searches | | Billing | 1 credit = 1 successful search request, which returns one page of results on any supported engine. A request that fails is not charged | | Account | Account-number login; email is optional; no KYC | | Payment | BTC, USDT, USDC, ETH, SOL, LTC, from a $25 top-up | ## What happened to Microsoft's Bing Search APIs? Microsoft retired the Bing Search APIs on 11 August 2025. Its lifecycle notice says existing instances were "decommissioned completely" and that new sign-ups closed. The notice was published in May 2025, so most teams had about three months to move. Microsoft's suggested replacement is **Grounding with Bing Search**, part of Azure AI Agents. It feeds web data into a language model's answer, so you receive a generated response with sources, not the list of results a person sees on bing.com. Microsoft's pricing page listed it at $14 per 1,000 transactions when we read it on 22 October 2026. PPC Land reported $35 per 1,000 in June 2025, against $25, $15 and $6 per 1,000 for the old S1, S2 and S3 Search API tiers. Our [Bing Search API replacement](https://serpcrypto.com/alternatives/bing-search-api/) page compares the migration options in detail. This page describes what the SerpCrypto Bing engine returns. ## What does engine=bing return? It returns the Bing web results page for your query as one JSON object. The structure matches the other engines, so code written for Google results reads Bing results with almost no change. | Key | Fields | |---|---| | `search_metadata` | `id`, `status`, `engine`, `created_at`, `credits_used` | | `search_parameters` | The query, country, language, location, device and page actually used | | `organic_results` | `position`, `title`, `link`, `displayed_link`, `snippet` | | `related_searches` | `query`, one item per suggestion at the bottom of the page | The `google_domain` parameter is ignored on this engine. A block that the page does not show is left out, so test for the key before you read it. The [response format](https://serpcrypto.com/docs/response/) page has a full example. *Python: the same query on Bing and on Google* ```python import os import requests API = "https://serpcrypto.com/api/v1/search" HEADERS = {"Authorization": "Bearer " + os.environ["SERPCRYPTO_API_KEY"]} def top_links(engine, query, n=10): params = {"engine": engine, "q": query, "gl": "us", "hl": "en"} resp = requests.get(API, headers=HEADERS, params=params, timeout=60) resp.raise_for_status() return [r["link"] for r in resp.json().get("organic_results", [])[:n]] bing = top_links("bing", "open source vector database") google = top_links("google", "open source vector database") shared = set(bing) & set(google) print("on both engines:", len(shared)) for link in bing: print("bing only" if link not in shared else "both", link) ``` ## Why would I track Bing results next to Google? Bing ranks pages with its own index and its own signals, so a site can sit 3rd on Bing and 15th on Google for the same query. Watching only one engine hides that gap. - **SEO reporting:** show clients their visibility on both engines with one job and one data format. - **AI products:** Microsoft builds Copilot on Bing, so Bing results give a view of the web that Microsoft's assistant starts from. - **Data quality:** two independent result lists make it easier to spot a missing or odd page in one of them. - **Coverage:** some regional and long-tail queries return different sources on Bing, which helps research and monitoring tools. ## How do I target a country, language or city on Bing? Use the same parameters as on Google: `gl` for the country, `hl` for the language, `location` for a city and `device` for desktop or mobile. You do not need to build a Bing market code such as `en-US` yourself. For example, `gl=ca&hl=fr` asks for French-language results as seen in Canada, and `location=Toronto, Ontario, Canada` narrows them to a city. Each combination is a separate search with its own credit. The [locations page](https://serpcrypto.com/docs/locations/) lists the codes for common markets. ## How does the SerpCrypto Bing engine differ from Grounding with Bing Search? Both use Bing, but they answer different needs. Facts on Microsoft's product come from its pricing and lifecycle pages, read on 22 October 2026. | | SerpCrypto engine=bing | Grounding with Bing Search | | --- | --- | --- | | What you receive | The results page as JSON: positions, links, snippets | A model answer grounded in web data, with sources | | Price | From $0.30/1K searches; $0.75/1K searches at the smallest pack | $14 per 1,000 transactions | | Account | Account number; no KYC, no card | Azure subscription, as part of Azure AI Agents | | Payment | Crypto: BTC, USDT, USDC, ETH, SOL, LTC | Azure billing | | Best for | Rank tracking, research, your own ranking or retrieval logic | Agents built on Azure that want Microsoft to do the grounding | *Prices checked on October 22, 2026.* ## How much does the Bing engine cost? Bing costs one credit per successful search, like every other engine. 10,000 searches cost $7.50 and 100,000 searches cost $56.25, as of October 2026. Credits never expire. See [all packs](https://serpcrypto.com/pricing/). | Plan | Quantity | Price | Price per unit | | --- | --- | --- | --- | | Test | 10,000 searches | $7.50 | $0.75/1K searches | | Starter | 25,000 searches | $18.75 | $0.75/1K searches | | Build | 100,000 searches | $56.25 | $0.56/1K searches | | Volume | 1,000,000 searches | $375 | $0.38/1K searches | In every plan: Every engine at 1 credit per search · Failed requests not charged · Credits never expire · Country, language, city and device targeting **Before you pay** - Minimum top-up $25. Top up once in crypto, then buy any plan from your balance. - Unused balance stays on your account and never expires, and it pays for search credits. When a purchase costs less than the minimum top-up, the part added only to reach the minimum is refundable on request to support within 30 days of the payment, in USDT or USDC, as long as it has not been spent. The rest of the balance is not paid back, except where the law requires it. - Credited once your payment is checked on the blockchain. Typical network confirmation time: BTC ≈ 20 min · USDT TRC-20 ≈ 1 min · USDT ERC-20 ≈ 5 min · USDC ERC-20 ≈ 5 min · ETH ≈ 5 min · SOL ≈ 1 min · LTC ≈ 15 min ## Bing Search API: common questions ### When was the Bing Search API retired? On 11 August 2025. Microsoft announced it in May 2025 and decommissioned every existing instance on that date. New sign-ups were closed. Its lifecycle page points users to Grounding with Bing Search in Azure AI Agents. ### Is this an official Microsoft API? No. SerpCrypto is an independent SERP API. It runs the search and returns the page Bing showed, parsed into JSON. It has no link with Microsoft and needs no Azure account. ### Can I get Bing News or Bing Images? Not at the moment. The Bing engine returns web results and related searches. For news and images, use `engine=google_news` or `engine=google_images` with the same key and the same credits. ### Do I have to rewrite my old Bing API code? Partly. The request becomes one GET with `engine=bing` and `q`, and the results move from Bing's `webPages.value` list to our `organic_results` array. The [migration guide](https://serpcrypto.com/alternatives/bing-search-api/) maps the old fields to the new ones. ### How do I pay without an Azure or company card? Top up from $25 in BTC, USDT, USDC, ETH, SOL, LTC, then buy a pack from your balance. SerpCrypto asks for no ID, no selfie and no company documents. Account-number login; email is optional. ## Sources 1. [Bing Search API retirement](https://learn.microsoft.com/en-us/lifecycle/announcements/bing-search-api-retirement) — Microsoft Learn (checked 2026-10-22) 2. [Grounding with Bing Search pricing](https://www.microsoft.com/en-us/bing/apis/grounding-pricing) — Microsoft (checked 2026-10-22) 3. [Microsoft ends Bing Search APIs on August 11; alternative costs 40-483% more](https://ppc.land/microsoft-ends-bing-search-apis-on-august-11-alternative-costs-40-483-more/) — PPC Land (2025-06-09) --- SerpCrypto: SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. Prepaid search credits from $0.30/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto. --- # Changelog > SerpCrypto changelog: dated changes to the SERP API, its fields, prices and policies, and how breaking changes are announced. Credits never expire. No KYC. - URL: https://serpcrypto.com/changelog/ - Last updated: 2026-10-22 - Publisher: SerpCrypto (https://serpcrypto.com) Every change that a customer or a parser could notice is written here with a date: new fields, engines, prices, policy updates and incidents. **Quick answer — What has changed at SerpCrypto, and when?** On 2026-10-22 **SerpCrypto** published its website, its API documentation for `https://serpcrypto.com/api/v1/search`, its price grid (from $7.50 for 10,000 searches down to $0.30/1K searches) and its policies. The API and the client area open on the date announced in this log. Every later change to fields, engines, prices or policies is recorded here. ## Change log | Date | Area | Change | | --- | --- | --- | | 2026-10-22 | Website | Website and API documentation published: quick start, search reference, response format, locations, errors, and code for Python, Node.js and curl. The API and the client area open on the date announced in this log. | | 2026-10-22 | Prices | First price grid published: seven credit packs from 10,000 searches ($7.50) to 5,000,000 searches ($1,500). Credits never expire. | | 2026-10-22 | Policies | Terms of service, privacy policy, refund policy and Acceptable Use Policy published. | ## What kinds of changes are logged? - **API fields:** a new block or field in the JSON, or a change in how an existing field is filled. - **Engines and parameters:** an engine or a parameter value added, renamed or retired. - **Prices:** any change to a pack price or size, with the date it applies. Packs already bought keep their credits. - **Policies:** terms, privacy, refunds and acceptable use, with a one-line summary of what changed. - **Incidents:** a period when an engine or the client area failed for many customers, and what was done. ## How are breaking changes to the API handled? The version is part of the path: `https://serpcrypto.com/api/v1`. Within this version, new fields can appear at any time, so a parser should ignore fields it does not know. A field that a parser may rely on is not renamed or removed within the version without a dated entry in this log first. Some changes come from the engines themselves. When Google stopped honouring `num=100` in September 2025, collecting 100 results started to take 10 requests instead of one. When an engine changes its page in a way that affects a block, the entry says which block and from which date. ## How do prices change? A new price applies to packs bought after the date in the log. It never reduces the credits already on an account, because credits are counted in searches, not in dollars. Prepaid search credits never expire: unused searches stay on your account until you use them. The current grid is always on the [pricing page](https://serpcrypto.com/pricing/) and in the machine-readable [pricing.json](https://serpcrypto.com/pricing.json) file. ## How can I follow this log? Check this page, or read its Markdown version, which carries the same table and is easy to diff. The [status page](https://serpcrypto.com/status/) explains how to test the API yourself when you suspect an incident. Questions about an entry go to support@serpcrypto.com; Email replies within 12 hours, 7 days a week, including help with API integration. --- SerpCrypto: SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. Prepaid search credits from $0.30/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto. --- # SERP API pricing compared: what 10,000, 100,000 and 1 million searches cost > SERP API pricing compared on 2026-10-22: what 10,000, 100,000 and 1 million searches a month cost at 15 APIs, with hidden costs, crypto and KYC. - URL: https://serpcrypto.com/compare/serp-api-pricing/ - Last updated: 2026-10-22 - Publisher: SerpCrypto (https://serpcrypto.com) Per-1,000 prices hide minimum plans, credit multipliers and expiry dates. This page turns public list prices into the monthly bill for three common volumes, then lists the costs that do not show on pricing pages. **Quick answer — How do SERP API prices compare in 2026?** On 2026-10-22, live Google results cost about $1.00 per 1,000 searches at 10,000 a month (Serper, Scrapeless), $0.75 at 100,000 (Serper, Serply) and $0.35 to $0.50 at 1 million (Scrapingdog's 5-credit Light search, Serper, Decodo). SerpApi's monthly plans cost $15.00 to $3.75 per 1,000 at those volumes. **SerpCrypto** charges $7.50, $56.25 and $375 for the same volumes, paid in crypto with no KYC. ## What does a month of Google searches cost at each provider? Monthly bill for 10,000, 100,000 and 1,000,000 Google searches (one results page each), using the least expensive self-serve plan, pack or pay-as-you-go rate that covers the volume. Prepaid packs are spread over the months they stay valid. | Provider | 10,000 a month | 100,000 a month | 1,000,000 a month | Unused credits | | --- | --- | --- | --- | --- | | SerpCrypto | $7.50 | $56.25 | $375 | Prepaid, never expires | | Serper | $10.00 | $75.00 | $500.00 | 6 months after purchase | | DataForSEO (Live) | $20.00 | $200.00 | $2,000 | Never | | DataForSEO (Standard queue) | $6.00 | $60.00 | $600.00 | Never | | Scrapingdog | $40.00 | $90.00 | $350.00 | Lite: end of cycle; Standard and up roll over | | Decodo Web Data API | $19.00 | $99.00 | $500.00 | Not stated | | Scrapeless | $10.00 | $100.00 | $1,000 | Unused monthly minimum not kept | | Serply | $15.00 | $75.00 | $750.00 | Purchased credits never expire | | Bright Data SERP API | $15.00 | $150.00 | $1,310 | Not stated | | Oxylabs Web Scraper API | $30.00 | $100.00 | — | Not stated | | Zenserp | $50.00 | $150.00 | $900.00 | Not stated | | SearchApi | $40.00 | $250.00 | $1,500 | Not stated | | ScrapingBee | $49.00 | $99.00 | $1,000 | No rollover | | HasData | $49.00 | $99.00 | — | End of period | | SerpApi | $150.00 | $725.00 | $3,750 | Monthly reset | *Prices checked on October 22, 2026.* List prices in USD, read on 2026-10-22. "—" = no self-serve option at that volume. DataForSEO's standard queue returns results after a short delay; every other line is live. Full method and 19 providers: [SERP API price index 2026](https://serpcrypto.com/data/serp-api-price-index-2026/). ## Which SERP API costs least at my volume? The answer changes with volume and with how fast you need results. These are the list prices on 2026-10-22: - **Under 10,000 searches a month:** DataForSEO's standard queue lists $0.60 per 1,000, with a $50 minimum payment and funds that never expire. For live results, Serper and Scrapeless list $1.00; Bright Data's recurring free tier of 5,000 requests a month covers very small needs. - **Around 100,000 a month:** Serper and Serply list $0.75 per 1,000 for live results, Scrapingdog $0.90, and Decodo, ScrapingBee and HasData $0.99. - **Around 1 million a month:** Scrapingdog lists $0.35 per 1,000 for its 5-credit Light search, and Serper and Decodo list about $0.50. SerpCrypto is in the first row of the table: $0.75/1K searches on 10,000 searches, $0.56/1K searches on 100,000 searches and $0.38/1K searches on 1,000,000 searches, prices set on 2026-10-22. We do not claim to be the cheapest at every volume. ## Which costs do pricing pages hide? Check six things before you compare two headline prices. 1. **Credits per Google search.** Several general scraping APIs charge many credits per search: 5 or 10 at Scrapingdog, 10 at ScrapingBee and SerpHouse, 25 at ScraperAPI. 2. **Expiry.** Monthly plans reset unused searches at renewal. Serper's packs last six months. Serply, DataForSEO and SerpCrypto keep purchased credits with no end date. 3. **Minimum plan size.** A $25 plan for 1,000 searches costs $25 per 1,000 however few you use. 4. **Rendering surcharges.** Oxylabs lists $1.00 per 1,000 without JavaScript rendering and $1.35 with it on its Starter plan. 5. **Depth.** Since September 2025, Google serves about 10 results per page, so a top-100 check costs 10 searches everywhere. 6. **Failed requests.** SerpApi does not count errored or failed searches, and neither does SerpCrypto. Check the policy of any API you shortlist. ## Which SERP APIs take crypto, and which ask for ID? Payment and identity rules stated on the providers' own pages. "Not mentioned" means the pages we read did not say. | Provider | Crypto payment | Identity checks | | --- | --- | --- | | SerpCrypto | BTC, USDT, USDC, ETH, SOL, LTC | None: no ID, no selfie, no documents | | Decodo Web Data API | Yes, self-serve (coins not stated) | Email verification | | Scrapeless | Yes (coins not stated) | None for basic packages | | DataForSEO | Bitcoin on request via support | None stated | | Bright Data SERP API | No | Not stated for the SERP API | | Oxylabs Web Scraper API | No | KYC questionnaire for every customer | | Serper | Not mentioned | None stated | | SerpApi | Not mentioned | None stated | | Scrapingdog | Not mentioned | None stated | *Prices checked on October 22, 2026.* ## Is a SERP API cheaper than an AI search API? Usually, per request. AI search APIs such as Brave, Perplexity, Linkup and You.com list about $5.00 per 1,000 queries, and Tavily $7.50 at entry and $5.00 at 100,000. They answer from their own index or retrieval system and return content prepared for a language model. A SERP API returns the results page of Google or Bing itself: the ranking, the ads, People Also Ask, the AI Overview and local results. Choose by output, then by price. The [AI agents guide](https://serpcrypto.com/use-cases/ai-agents/) explains when raw Google results are the better input. ## What does SerpCrypto cost? Prepaid packs, one credit per successful search on any engine, credits that never expire. Prices set on 2026-10-22. | Plan | Quantity | Price | Price per unit | | --- | --- | --- | --- | | Test | 10,000 searches | $7.50 | $0.75/1K searches | | Build | 100,000 searches | $56.25 | $0.56/1K searches | | Volume | 1,000,000 searches | $375 | $0.38/1K searches | | Bulk | 5,000,000 searches | $1,500 | $0.30/1K searches | In every plan: Every engine at 1 credit per search · Failed requests not charged · Credits never expire · Country, language, city and device targeting **Before you pay** - Minimum top-up $25. Top up once in crypto, then buy any plan from your balance. - Unused balance stays on your account and never expires, and it pays for search credits. When a purchase costs less than the minimum top-up, the part added only to reach the minimum is refundable on request to support within 30 days of the payment, in USDT or USDC, as long as it has not been spent. The rest of the balance is not paid back, except where the law requires it. - Credited once your payment is checked on the blockchain. Typical network confirmation time: BTC ≈ 20 min · USDT TRC-20 ≈ 1 min · USDT ERC-20 ≈ 5 min · USDC ERC-20 ≈ 5 min · ETH ≈ 5 min · SOL ≈ 1 min · LTC ≈ 15 min ## SERP API pricing: common questions ### What is the cheapest SERP API in 2026? It depends on volume and speed. On 2026-10-22, DataForSEO's standard queue listed $0.60 per 1,000 for delayed results. For live results, Serper and Scrapeless listed $1.00 at 10,000 searches a month, Serper and Serply $0.75 at 100,000, and Scrapingdog's 5-credit Light search $0.35 at 1 million. ### Why is SerpApi more expensive than Serper? SerpApi sells monthly plans starting at $25 for 1,000 searches and includes many engines, hourly throughput guarantees and, from its Production plan, a U.S. Legal Shield of up to $2 million. Serper focuses on Google and lists $1.00 per 1,000 at entry, falling to $0.30 at 12.5 million. ### Do SERP API credits expire? At most providers, yes: monthly plans reset at renewal and Serper's packs expire after six months. DataForSEO funds, Serply purchased credits and SerpCrypto credits do not expire. ### How much does rank tracking cost after num=100 was removed? About ten times more for deep checks: a top-100 position check now takes 10 searches. At $0.75 per 1,000, checking 1,000 keywords to depth 100 every week costs about $30 a month; to depth 10, about $3. ### Can I pay for a SERP API with crypto? Yes, with a few. Decodo and Scrapeless state self-serve crypto payment, DataForSEO takes Bitcoin on request, and SerpCrypto takes BTC, USDT, USDC, ETH, SOL, LTC with no KYC. Bright Data and Oxylabs say they do not accept crypto. ## Sources 1. [Serper: pricing](https://serper.dev/) — Serper (read 2026-10-22) 2. [DataForSEO: pricing](https://dataforseo.com/apis/serp-api/pricing) — DataForSEO (read 2026-10-22) 3. [Scrapingdog: pricing](https://www.scrapingdog.com/pricing/) — Scrapingdog (read 2026-10-22) 4. [Decodo Web Data API: pricing](https://decodo.com/pricing) — Decodo (read 2026-10-22) 5. [Scrapeless: pricing](https://www.scrapeless.com/en/pricing) — Scrapeless (read 2026-10-22) 6. [Serply: pricing](https://serply.io/pricing) — Serply (read 2026-10-22) 7. [Bright Data SERP API: pricing](https://www.bright.cn/pricing/serp) — Bright Data (read 2026-10-22) 8. [Oxylabs Web Scraper API: pricing](https://oxylabs.io/pricing/web-scraper-api) — Oxylabs (read 2026-10-22) 9. [Zenserp: pricing](https://zenserp.com/pricing-plans/) — Zenserp (read 2026-10-22) 10. [SearchApi: pricing](https://www.searchapi.io/pricing) — SearchApi (read 2026-10-22) 11. [ScrapingBee: pricing](https://www.scrapingbee.com/documentation/google-api/) — ScrapingBee (read 2026-10-22) 12. [HasData: pricing](https://hasdata.com/prices) — HasData (read 2026-10-22) 13. [SerpApi: pricing](https://serpapi.com/pricing) — SerpApi (read 2026-10-22) 14. [Tavily API credits](https://docs.tavily.com/documentation/api-credits) — Tavily (read 2026-10-22) 15. [Brave Search API](https://brave.com/search/api/) — Brave (read 2026-10-22) --- SerpCrypto: SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. Prepaid search credits from $0.30/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto. --- # SerpApi vs DataForSEO (2026): two very different ways to buy Google results > SerpApi vs DataForSEO in 2026: monthly plans vs pay-as-you-go queues, price per 1,000 searches, extras, payment and minimums, checked on 22 October 2026. - URL: https://serpcrypto.com/compare/serpapi-vs-dataforseo/ - Last updated: 2026-10-22 - Publisher: SerpCrypto (https://serpcrypto.com) SerpApi sells real-time searches on monthly plans. DataForSEO sells results per page from a prepaid balance, at three speeds. This page compares prices, extras and payment terms. It is published by SerpCrypto, a competitor of both, and our column is labelled. **Quick answer — SerpApi or DataForSEO: which is better for Google results in 2026?** DataForSEO costs less if you can wait: $0.60 per 1,000 results pages in its standard queue (about 5 minutes on average), $2.00 in live mode, from a $50 balance that never expires (22 October 2026). SerpApi answers in real time with about 125 APIs and a legal shield, at $7.25 per 1,000 on its 100,000-search plan. **SerpCrypto**, this page's publisher, adds crypto and no KYC. ## SerpApi vs DataForSEO vs SerpCrypto: the facts side by side Competitor facts come from their own pages, read on 22 October 2026. The last column is ours. | | SerpApi | DataForSEO SERP API | SerpCrypto (ours) | | --- | --- | --- | --- | | How you buy | Monthly plans, month to month | Pay as you go from a balance | Prepaid packs from a crypto balance | | Speed options | Real time | Standard queue (about 5 min), priority (up to 1 min), live (up to 6 s), per DataForSEO | Real time | | Price per 1,000 results pages | $25.00 (Starter) to $1.96 (largest Cloud plan) | $0.60 standard, $1.20 priority, $2.00 live | $0.75/1K searches to $0.30/1K searches | | Unused value | Searches reset each billing cycle | "The funds on your account never expire" | Credits never expire | | Minimum | Starter at $25 a month | $50 minimum payment | $25 top-up | | Free start | 250 searches a month | $1 credit at sign-up | None | | Billing unit | 1 search, whatever the number of results | Per page of up to 10 results; depth 100 costs 10 times depth 10 | 1 credit per page of results | | Extras | AI Overview may need an extra request (billing not stated) | Async AI Overview +$0.0006; People Also Ask clicks $0.00015 each; search operators ×5 | None: AI Overview included when shown; operators count as one search | | Legal cover | U.S. Legal Shield, up to $2 million, Production and above | None stated | None | | Payment | Credit cards (Stripe) | PayPal, Amazon Pay, cards under $5,000; wire or ACH from $5,000; Bitcoin on request | BTC, USDT, USDC, ETH, SOL, LTC | | Identity checks | None stated | None stated | No KYC | *Prices checked on October 22, 2026.* Sources at the bottom of the page. We read only DataForSEO's Google SERP pages. ## How much does each cost for a steady monthly volume? Effective price per 1,000 Google results pages. SerpApi: the least expensive plan that covers the volume. DataForSEO: its per-page price. SerpCrypto: the pack equal to one month, at the prices on this page, set on 2026-10-22. | Pages per month | SerpApi | DataForSEO standard | DataForSEO live | SerpCrypto (ours) | | --- | --- | --- | --- | --- | | 1,000 | $25.00 | $0.60 ($50 minimum payment) | $2.00 | $0.75/1K searches (10,000 searches lasts 10 months) | | 10,000 | $15.00 | $0.60 | $2.00 | $0.75/1K searches | | 100,000 | $7.25 | $0.60 | $2.00 | $0.56/1K searches | | 1,000,000 | $3.75 | $0.60 | $2.00 | $0.38/1K searches | *Prices checked on October 22, 2026.* DataForSEO's standard queue is asynchronous: you post a task and collect the result later. Live mode is the closest to a real-time API. ## What does DataForSEO's queue mean in practice? It means you trade waiting time for price. DataForSEO states three modes for its Google SERP API: a standard queue at $0.0006 per page with a turnaround of "5 minutes on average", a priority queue at $0.0012 with "up to 1 minute on average", and live mode at $0.002 with "up to 6 seconds on average". For a nightly rank-tracking job, a 5-minute delay rarely matters. For an AI agent or a user waiting on a screen, only live mode fits, and live mode costs $2.00 per 1,000 pages. ## Which extras change the bill? DataForSEO prices several options on top of the base page. Its task documentation lists an extra $0.0006 for loading an asynchronous AI Overview, $0.00015 for each People Also Ask click, and a charge multiplied by 5 for queries that use search operators such as `site:` or `intitle:`. Since 19 September 2025 it bills per 10 results, so depth 100 costs 10 times depth 10. SerpApi counts one search whatever the number of results, and does not count cached, errored or failed searches. It notes that Google sometimes requires an additional request to retrieve AI Overview results; whether that request is billed is not stated on the page we read. On SerpCrypto, a page of results is one credit, with operators in the query and the AI Overview when Google shows one. ## Which should I choose for my use case? - **Large SEO datasets that can wait minutes:** DataForSEO's standard queue, at $0.60 per 1,000 pages. - **Real time with many engines and a vendor that assumes collection liability:** SerpApi, from its Production plan. - **Live Google results at a steady volume, paid by card or PayPal:** compare DataForSEO live at $2.00 with the SerpApi plan for your volume. - **Real-time Google and Bing results, paid in crypto, with no KYC:** SerpCrypto, the publisher of this page. DataForSEO's balance and SerpCrypto's credits share one trait: neither expires. SerpApi's monthly searches reset each billing cycle. ## Can I pay DataForSEO or SerpApi in crypto? DataForSEO accepts Bitcoin on request. Its help centre says: "Please contact our 24/7 support team if you want to pay in Bitcoin". Its usual methods are PayPal, Amazon Pay and cards for purchases under $5,000, with wire transfer or ACH from $5,000. SerpApi's legal page references credit cards, processed through Stripe, and no other method. SerpCrypto takes only crypto: BTC, USDT, USDC, ETH, SOL, LTC, with USDT on TRC-20 or ERC-20, from a $25 top-up, and asks for no ID. ## What does SerpCrypto cost? Real-time searches from $7.50 for 10,000 searches to $375 for 1,000,000 searches, one credit per page on every engine, credits that never expire. | Plan | Quantity | Price | Price per unit | | --- | --- | --- | --- | | Test | 10,000 searches | $7.50 | $0.75/1K searches | | Build | 100,000 searches | $56.25 | $0.56/1K searches | | Scale | 500,000 searches | $281.25 | $0.56/1K searches | | Volume | 1,000,000 searches | $375 | $0.38/1K searches | In every plan: Every engine at 1 credit per search · Failed requests not charged · Credits never expire · Country, language, city and device targeting **Before you pay** - Minimum top-up $25. Top up once in crypto, then buy any plan from your balance. - Unused balance stays on your account and never expires, and it pays for search credits. When a purchase costs less than the minimum top-up, the part added only to reach the minimum is refundable on request to support within 30 days of the payment, in USDT or USDC, as long as it has not been spent. The rest of the balance is not paid back, except where the law requires it. - Credited once your payment is checked on the blockchain. Typical network confirmation time: BTC ≈ 20 min · USDT TRC-20 ≈ 1 min · USDT ERC-20 ≈ 5 min · USDC ERC-20 ≈ 5 min · ETH ≈ 5 min · SOL ≈ 1 min · LTC ≈ 15 min ## SerpApi vs DataForSEO: common questions ### Is DataForSEO cheaper than SerpApi? Yes, at the list prices read on 22 October 2026. DataForSEO's standard queue costs $0.60 per 1,000 pages and its live mode $2.00. SerpApi costs $7.25 per 1,000 on its 100,000-search plan and $3.75 on its 1 million plan. ### Is DataForSEO real time? Only in live mode, which DataForSEO describes as "up to 6 seconds on average" at $0.002 per page. The standard queue averages about 5 minutes and the priority queue up to 1 minute. ### Does DataForSEO charge more for 100 results? Yes. Since 19 September 2025, it bills per page of up to 10 results, so a depth of 100 costs 10 times a depth of 10: $0.006 instead of $0.0006 in the standard queue. ### Do DataForSEO funds expire? No. Its help centre says the funds on your account never expire. The minimum payment is $50. SerpCrypto credits do not expire either; SerpApi searches reset each billing cycle. ### Who wrote this comparison? SerpCrypto, a SERP API that competes with both. Every competitor fact comes from the provider's own pages, listed below. Prices change, so check them before you buy. ## Sources 1. [SerpApi pricing](https://serpapi.com/pricing) — SerpApi (read 2026-10-22) 2. [SerpApi legal](https://serpapi.com/legal) — SerpApi (read 2026-10-22) 3. [SerpApi: Google AI Overview](https://serpapi.com/ai-overview) — SerpApi (read 2026-10-22) 4. [Google SERP API](https://dataforseo.com/apis/serp-api/google-serp-api) — DataForSEO (read 2026-10-22) 5. [Google Organic SERP: task_post](https://docs.dataforseo.com/v3/serp/google/organic/task_post/) — DataForSEO (read 2026-10-22) 6. [Critical updates: depth](https://dataforseo.com/update/google-organic-serp-api-critical-updates-depth) — DataForSEO (read 2026-10-22) 7. [Making a payment, changing details, refunds](https://dataforseo.com/help-center/making-a-payment-changing-details-refunds) — DataForSEO (read 2026-10-22) 8. [How does your free trial work?](https://dataforseo.com/help-center/how-does-your-free-unlimited-trial-work) — DataForSEO (read 2026-10-22) --- SerpCrypto: SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. Prepaid search credits from $0.30/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto. --- # SerpApi vs Serper (2026): price, credits and engines compared > SerpApi vs Serper in 2026: price per 1,000 searches at 4 volumes, free tiers, credit expiry, engines, legal shield and payment, checked on 22 October 2026. - URL: https://serpcrypto.com/compare/serpapi-vs-serper/ - Last updated: 2026-10-22 - Publisher: SerpCrypto (https://serpcrypto.com) SerpApi is the broad, premium option; Serper is the low-cost Google specialist. This page compares their public plans line by line and says which one fits which job. It is published by SerpCrypto, a competitor of both, and our own column is labelled. **Quick answer — SerpApi or Serper: which SERP API should I use in 2026?** Choose Serper for low-cost Google results: $1.00 per 1,000 queries on its $50 pack, down to $0.30 at 12.5 million, credits valid 6 months (22 October 2026). Choose SerpApi for breadth and legal cover: about 125 APIs and a U.S. Legal Shield, but $7.25 per 1,000 at 100,000 searches a month. **SerpCrypto**, the publisher of this page, adds crypto payment and no KYC. ## SerpApi vs Serper vs SerpCrypto: the facts side by side Competitor facts come from their own pages, read on 22 October 2026. The last column is ours. | | SerpApi | Serper | SerpCrypto (ours) | | --- | --- | --- | --- | | Billing model | Monthly plans | Prepaid packs | Prepaid packs | | Unused searches | "reset at the start of the new billing cycle" | Credits valid for 6 months | Never expire | | Free start | 250 searches a month | 2,500 queries once, no credit card | None | | Smallest paid option | Starter: $25 for 1,000 a month | $50 for 50,000 | $7.50 for 10,000 searches | | What counts | Successful searches; "Cached, errored, and failed searches are not" | Credits deducted for successful responses | 1 credit per successful search; failed requests free | | Engines | About 125 APIs: some 60 Google variants (incl. AI Overview and AI Mode), Bing, Baidu, Yandex, Yahoo!, DuckDuckGo, Naver, YouTube, Amazon, Walmart, eBay and more | Google only: Search, Images, News, Maps, Places, Videos, Shopping, Scholar, Patents, Autocomplete | Google Search, News, Images, Maps, Shopping and Bing | | Throughput | Per plan per hour, e.g. 20,000 on Searcher | 50 to 300 queries per second by pack | Concurrency limit in the dashboard | | Speed | "Each API request runs immediately"; no figure | "1-2 seconds"; about 0.6–0.7 s median in Proxyway's January 2026 tests | Not published until measured | | Legal cover | U.S. Legal Shield, up to $2 million, Production plan and above | None stated | None | | Payment | Credit cards (Stripe) | Visa, Mastercard, American Express, Discover, PayPal | BTC, USDT, USDC, ETH, SOL, LTC | | Identity checks | None stated | None stated | No KYC | | Refund | Full refund within 7 days unless more than 20% used | Within 7 days of first payment if less than 20% used | Within 72 hours if less than 20% used, to your balance | | MCP | Hosted MCP server, MIT licence | — | Copy-paste MCP server code in the docs | *Prices checked on October 22, 2026.* "—" = not stated on the pages we read. Sources at the bottom of the page. ## How much does each cost per 1,000 searches? Effective price per 1,000 Google searches for a steady monthly volume. SerpApi: the least expensive plan that covers the volume. Serper: the least expensive pack that covers it within its 6-month validity. SerpCrypto: the pack equal to one month of searches, at the prices on this page, set on 2026-10-22. | Searches per month | SerpApi | Serper | SerpCrypto (ours) | | --- | --- | --- | --- | | 1,000 | $25.00 (Starter) | $8.33 ($50 pack used over 6 months) | $0.75/1K searches (10,000 searches lasts 10 months) | | 10,000 | $15.00 (Production, $150) | $1.00 ($50 pack lasts 5 months) | $0.75/1K searches | | 100,000 | $7.25 (Searcher, $725) | $0.75 ($375 pack lasts 5 months) | $0.56/1K searches | | 1,000,000 | $3.75 (Cloud 1M, $3,750) | $0.50 ($1,250 pack lasts 2.5 months) | $0.38/1K searches | *Prices checked on October 22, 2026.* Since SerpCrypto credits do not expire, a larger pack lowers the rate further. Prices exclude taxes. ## Why is SerpApi so much more expensive than Serper? SerpApi charges for breadth and guarantees, Serper for volume. SerpApi's plans cover about 125 APIs, from Google's many verticals to Amazon, YouTube and Yelp, and from the Production plan up they include a U.S. Legal Shield of up to $2 million: SerpApi says it assumes liability "for the lawful collection of public search data". Each plan also states a guaranteed number of successful searches per hour. Serper sells Google only, as prepaid packs, with no legal cover stated. Its list price is $1.00 per 1,000 at the entry pack and $0.30 at 12.5 million. Proxyway's report of 4 March 2026 put the two at $25 to $7.25 and $1 to $0.75 per 1,000 for the plans it tested. ## Which one is faster? Serper has the published edge. In Proxyway's 2026 report, based on three rounds of 2,000 requests in mid-January 2026, Serper and Oxylabs were the fastest real-time APIs, at about 0.6 to 0.7 seconds at the median. Serper's own page says "Google search results in 1-2 seconds". SerpApi publishes no latency figure; its homepage says "Each API request runs immediately – no waiting for results." Speed also depends on the engine, the location and the features you ask for, so test both on your own queries before you decide. ## Which should I choose for my use case? - **Rank tracking on Google at volume:** Serper, for its per-1,000 rate, unless you need a legal cover or other engines. - **Many engines in one contract** (Google, Bing, Baidu, YouTube, Amazon): SerpApi. - **A company that wants a vendor to assume collection liability:** SerpApi, from its Production plan. - **A first test without a card:** Serper, with 2,500 free queries. - **AI agents that need Google AI Mode:** SerpApi, which lists a Google AI Mode API. - **Crypto payment, no KYC, credits that never expire, Google and Bing at one price:** SerpCrypto, the publisher of this page. ## What about the lawsuit against SerpApi? Google sued SerpApi in December 2025 under the anti-circumvention rule of the US Digital Millennium Copyright Act. On 20 July 2026 the court dismissed the complaint; claims about plain search results were dismissed without leave to amend. Google filed an amended complaint on 10 August 2026, based on content-licensing terms. These are allegations; no ruling against SerpApi had been made when we checked. The case concerns how search results are collected, which is the business of every SERP API, Serper and SerpCrypto included. Our [legal overview](https://serpcrypto.com/guides/is-a-serp-api-legal/) gives the timeline; it is not legal advice. ## What does SerpCrypto cost? Prepaid packs from $7.50 to $1,500, one credit per successful search on every engine, credits that never expire. Paid in BTC, USDT, USDC, ETH, SOL, LTC. | Plan | Quantity | Price | Price per unit | | --- | --- | --- | --- | | Test | 10,000 searches | $7.50 | $0.75/1K searches | | Build | 100,000 searches | $56.25 | $0.56/1K searches | | Volume | 1,000,000 searches | $375 | $0.38/1K searches | | Bulk | 5,000,000 searches | $1,500 | $0.30/1K searches | In every plan: Every engine at 1 credit per search · Failed requests not charged · Credits never expire · Country, language, city and device targeting **Before you pay** - Minimum top-up $25. Top up once in crypto, then buy any plan from your balance. - Unused balance stays on your account and never expires, and it pays for search credits. When a purchase costs less than the minimum top-up, the part added only to reach the minimum is refundable on request to support within 30 days of the payment, in USDT or USDC, as long as it has not been spent. The rest of the balance is not paid back, except where the law requires it. - Credited once your payment is checked on the blockchain. Typical network confirmation time: BTC ≈ 20 min · USDT TRC-20 ≈ 1 min · USDT ERC-20 ≈ 5 min · USDC ERC-20 ≈ 5 min · ETH ≈ 5 min · SOL ≈ 1 min · LTC ≈ 15 min ## SerpApi vs Serper: common questions ### Is Serper cheaper than SerpApi? Yes, by a wide margin at the list prices read on 22 October 2026. At 100,000 searches a month, SerpApi's Searcher plan costs $725, or $7.25 per 1,000, while Serper's $375 pack for 500,000 queries works out at $0.75 per 1,000. ### Do SerpApi and Serper credits roll over? SerpApi's searches reset at the start of each billing cycle; on a downgrade, remaining searches move to an "Extra Credits" balance. Serper's prepaid credits are valid for 6 months. SerpCrypto credits never expire. ### Which one supports Bing? SerpApi does; Serper covers Google only. SerpCrypto also returns Bing with `engine=bing`, for one credit per search. ### Can I pay SerpApi or Serper in crypto? Neither lists crypto on the pages we read on 22 October 2026: SerpApi references credit cards through Stripe, and Serper lists cards and PayPal. SerpCrypto takes BTC, USDT, USDC, ETH, SOL, LTC. ### Who wrote this comparison? SerpCrypto, a SERP API that competes with both. We used each provider's own pages and one independent report, all listed below, and labelled our own column. Check the providers' pages before you buy: prices change. ## Sources 1. [SerpApi pricing](https://serpapi.com/pricing) — SerpApi (read 2026-10-22) 2. [SerpApi homepage](https://serpapi.com/) — SerpApi (read 2026-10-22) 3. [SerpApi legal](https://serpapi.com/legal) — SerpApi (read 2026-10-22) 4. [SerpApi MCP server](https://github.com/serpapi/serpapi-mcp) — SerpApi on GitHub (read 2026-10-22) 5. [Google AI Mode API](https://serpapi.com/google-ai-mode-api) — SerpApi (read 2026-10-22) 6. [Serper: Google Search API](https://serper.dev/) — Serper (read 2026-10-22) 7. [Serper terms of service](https://serper.dev/terms) — Serper (read 2026-10-22) 8. [Search APIs 2026 report](https://proxyway.com/research/search-apis-2026-report) — Proxyway (2026-03-04, read 2026-10-22) 9. [Court dismisses Google's DMCA claims against SerpApi](https://www.searchenginejournal.com/court-dismisses-googles-dmca-claims-against-serpapi/583033/) — Search Engine Journal (2026-07-21, read 2026-10-22) 10. [Google amends SerpApi suit with content licensing terms](https://www.searchenginejournal.com/google-amends-serpapi-suit-with-content-licensing-terms/585505/) — Search Engine Journal (read 2026-10-22) --- SerpCrypto: SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. Prepaid search credits from $0.30/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto. --- # Contact SerpCrypto > Contact SerpCrypto: API and billing support by email, replies within 12 hours, plus a separate abuse address. What to include, and what we never ask for. - URL: https://serpcrypto.com/contact/ - Last updated: 2026-10-22 - Publisher: SerpCrypto (https://serpcrypto.com) Two addresses, each read by a person. Include the details below and most questions are solved on the first reply. **Quick answer — How do I contact SerpCrypto?** Email support@serpcrypto.com for the API, payments and your account: Email replies within 12 hours, 7 days a week, including help with API integration. Report abuse to abuse@serpcrypto.com. **SerpCrypto** is run by SerpGoat LLC. We never ask for your account number or your API key by email or chat. ## Addresses | Attribute | Value | | --- | --- | | API, payments, account | support@serpcrypto.com | | Abuse reports | abuse@serpcrypto.com | | Reply time | Email replies within 12 hours, 7 days a week, including help with API integration | | Languages | English | | Operator | SerpGoat LLC | ## What details help support answer on the first reply? - **API problem:** the request ID from `search_metadata.id`, the time with your time zone, the parameters you sent (without your API key) and the HTTP status you received. The [error reference](https://serpcrypto.com/docs/errors/) may already have the fix. - **Payment not credited:** the coin, the network, the amount and the transaction ID (TXID). Payments are credited once confirmed on the blockchain; the [crypto payments page](https://serpcrypto.com/crypto-payments/) lists the usual confirmation counts. - **Refund:** your order ID and which part of the [refund policy](https://serpcrypto.com/refunds/) applies. - **Higher concurrency:** your expected searches per minute and the engines you use. ## What will we never ask you for? We never ask for your account number or your API key by email or chat. We will not ask for a password, a seed phrase, a private key or identity documents either. A message that asks for any of these does not come from us: do not answer it, and forward it to support@serpcrypto.com. ## How do I report abuse? Write to abuse@serpcrypto.com with what you saw, when, and any request ID or evidence. Abuse reports are reviewed by a person, and confirmed abuse leads to suspension. The [Acceptable Use Policy](https://serpcrypto.com/acceptable-use/) lists what is banned. --- SerpCrypto: SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. Prepaid search credits from $0.30/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto. --- # How to pay for SERP API credits with crypto > How to pay for a SERP API with crypto: top up from $25 in BTC, USDT, USDC, ETH, SOL or LTC, then buy search credits. No KYC, no card needed. - URL: https://serpcrypto.com/crypto-payments/ - Last updated: 2026-10-22 - Publisher: SerpCrypto (https://serpcrypto.com) You top up a balance once, on the network of your choice, then buy search credit packs from it. This page explains each step, which coin to pick and what to do if something goes wrong. **Quick answer — How do I pay for a SERP API with crypto?** On **SerpCrypto**, you top up your balance with at least $25 in BTC, USDT, USDC, ETH, SOL, LTC, sending the exact amount shown to the address shown, on the listed network. The top-up is credited once the payment is checked on the blockchain; you then buy search credits, from $7.50 for 10,000 searches. No KYC, no card. ## Which cryptocurrencies can I pay with? These coins and networks are accepted, and only these. USDT works on Tron (TRC-20) or Ethereum (ERC-20); USDC on Ethereum (ERC-20). - Bitcoin (BTC) - Tether (USDT) — TRC-20 or ERC-20 - USD Coin (USDC) — ERC-20 - Ethereum (ETH) - Solana (SOL) - Litecoin (LTC) ## How payment works 1. **Top up $25 or more** — Choose the amount and the coin (BTC, USDT, USDC, ETH, SOL, LTC). The crypto amount and its rate are locked for 30 minutes. 2. **Send it to the address shown, on the exact network** — The payment page shows our deposit address for the coin and network you picked: BTC on Bitcoin, USDT on Tron (TRC-20) or Ethereum (ERC-20), USDC on Ethereum (ERC-20), ETH on Ethereum, SOL on Solana, LTC on Litecoin. Send only that coin on that network: funds sent on another network or in another coin can’t be recovered. The network fee comes on top of the amount. 3. **Send the exact amount shown, then click “I’ve sent the payment”** — Each top-up has its own exact amount (down to the last decimals): that is how we find your payment and link it to your account. 4. **Credited once your payment is checked on the blockchain** — Your balance is credited once we have checked the transaction on the blockchain, after the network has confirmed it. Typical network confirmation times: 5. **One balance for all your search packs** — Starting from a pack, you top up its exact price (or the minimum for a small pack): the pack is bought when the top-up is credited and your API key shows in the dashboard. Anything left on the balance buys your next pack. Unused balance stays on your account and never expires, and it pays for search credits. When a purchase costs less than the minimum top-up, the part added only to reach the minimum is refundable on request to support within 30 days of the payment, in USDT or USDC, as long as it has not been spent. The rest of the balance is not paid back, except where the law requires it. | Crypto | Network | Confirmations | Usual time | | --- | --- | --- | --- | | Bitcoin (BTC) | Bitcoin | 2 | ≈ 20 min | | Tether (USDT) | Tron (TRC-20) | 20 | ≈ 1 min | | Tether (USDT) | Ethereum (ERC-20) | 12 | ≈ 5 min | | USD Coin (USDC) | Ethereum (ERC-20) | 12 | ≈ 5 min | | Ethereum (ETH) | Ethereum | 12 | ≈ 5 min | | Solana (SOL) | Solana | 1 | ≈ 1 min | | Litecoin (LTC) | Litecoin | 6 | ≈ 15 min | Closed the tab? The pending top-up stays under Billing in your account: open it to see the address and the exact amount again. ## Which coin and network should I choose? Choose the coin you already hold, on a network your wallet or exchange supports for withdrawals. Three practical points help: - **Stablecoins keep the dollar value steady.** USDT and USDC are pegged to the US dollar, so the amount you send matches the amount credited, without a price move during confirmation. - **Network fees differ a lot.** Fees are paid to the network, not to us, and change with traffic. Check the fee your wallet shows before you confirm. - **Confirmation times differ too.** The table above lists how many confirmations each network needs and the usual waiting time. If you have no preference, a stablecoin on a network your wallet already uses is the simplest choice. ## Why is there a minimum top-up? Each top-up has to be found and checked on the blockchain, so very small payments cost more to handle than they bring. The minimum is $25. If the pack you want costs less than that, for example 10,000 searches at $7.50, the rest stays on your balance for later packs. Unused balance stays on your account and never expires, and it pays for search credits. When a purchase costs less than the minimum top-up, the part added only to reach the minimum is refundable on request to support within 30 days of the payment, in USDT or USDC, as long as it has not been spent. The rest of the balance is not paid back, except where the law requires it. ## What happens after I send the payment? Your top-up is matched by its exact amount and credited once the transaction is confirmed and checked on the blockchain. Then you buy a credit pack from your balance, and the credits and your API key appear in your dashboard. 1. Send exactly the amount shown, on the network shown. 2. Click "I've sent the payment" and keep the transaction ID (TXID) your wallet gives you. 3. Wait for the confirmations listed in the table above. 4. Buy a pack. Your API key appears in your dashboard as soon as you buy a credit pack from your confirmed crypto balance. ## What if the amount or the network is wrong? Send only on a network listed on the payment page. Coins sent on another network, or to an address for another coin, may not be recoverable. - **Wrong amount:** contact support at support@serpcrypto.com with the TXID and your order ID. We check it by hand. - **Wrong network or wrong coin:** contact support with the TXID. Recovery depends on the network and is not always possible. - **Payment not visible after the usual time:** check the TXID on a block explorer. If it is confirmed, send it to support. We never ask for your account number or your API key by email or chat. ## Can I get a refund in crypto? Refunds follow our [refund policy](https://serpcrypto.com/refunds/). Within 72 hours of purchase, a credit pack is credited back to your balance if you used less than 20% of its searches. A pack that does not work, and that support cannot fix within 24 hours, is credited back in full. Top-ups are credited to your balance and are not paid back, except where the law requires it; the part added only to reach the minimum top-up is refundable on request within 30 days, in USDT or USDC, if unspent. In short, an unused pack can go back to your balance within the stated window, and only the part added to reach the minimum top-up can be paid back in crypto, in USDT or USDC. ## What can I buy once my balance is credited? Any credit pack, paid from your balance: 10,000 searches for $7.50, 25,000 searches for $18.75, 100,000 searches for $56.25, as of October 2026. See [all packs](https://serpcrypto.com/pricing/). | Plan | Quantity | Price | Price per unit | | --- | --- | --- | --- | | Test | 10,000 searches | $7.50 | $0.75/1K searches | | Starter | 25,000 searches | $18.75 | $0.75/1K searches | | Build | 100,000 searches | $56.25 | $0.56/1K searches | In every plan: Every engine at 1 credit per search · Failed requests not charged · Credits never expire · Country, language, city and device targeting **Before you pay** - Minimum top-up $25. Top up once in crypto, then buy any plan from your balance. How payment works ## Paying with crypto: common questions ### Can I pay with Monero or another coin? No. SerpCrypto accepts BTC, USDT, USDC, ETH, SOL, LTC only, on the networks listed on this page. Coins and networks not listed are not credited. ### Do you accept cards or bank transfers? No. SerpCrypto is crypto only. If you buy crypto on an exchange first, the exchange may ask for its own identity checks; SerpCrypto does not. ### Do other SERP APIs accept crypto? A few do. When we read their pages on 22 October 2026, Decodo and Scrapeless stated that they accept cryptocurrency without naming the coins, and DataForSEO took Bitcoin on request through support. SerpCrypto lists its six coins and networks up front. See the [comparison on our no-KYC page](https://serpcrypto.com/no-kyc-serp-api/). ### Do I need to verify my identity to pay? No. Paying needs no identity check of any kind: no ID upload, no selfie, no paperwork. Account-number login; email is optional. See [no-KYC SERP API](https://serpcrypto.com/no-kyc-serp-api/). ### Does my balance expire? No. Unused balance stays on your account, and search credits bought from it never expire either. Prepaid search credits never expire: unused searches stay on your account until you use them. ### Can I get an invoice? Your dashboard lists your top-ups and orders. For any other paperwork, write to support@serpcrypto.com. ## Sources 1. [Pricing](https://decodo.com/pricing) — Decodo (checked 2026-10-22) 2. [Pricing](https://www.scrapeless.com/en/pricing) — Scrapeless (checked 2026-10-22) 3. [Making a payment, changing details, refunds](https://dataforseo.com/help-center/making-a-payment-changing-details-refunds) — DataForSEO Help Center (checked 2026-10-22) --- SerpCrypto: SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. Prepaid search credits from $0.30/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto. --- # SERP API price index 2026: what 1,000 searches cost at 19 providers > SERP API price index 2026: list prices of 19 providers per 1,000 searches at 4 volumes, free tiers, credit expiry, crypto and KYC. Checked 2026-10-22. - URL: https://serpcrypto.com/data/serp-api-price-index-2026/ - Last updated: 2026-10-22 - Publisher: SerpCrypto (https://serpcrypto.com) The public list prices of 19 Google SERP APIs, normalised to the cost of 1,000 searches at four monthly volumes, with free tiers, credit expiry, crypto payment and identity checks. Published by SerpCrypto, which is one of the APIs in the table. **Quick answer — How much does a SERP API cost per 1,000 searches in 2026?** Across 19 Google SERP APIs whose list prices we read on 2026-10-22, the median entry price is $2.00 per 1,000 searches, from $0.60 to $25.00. Buying 10,000 searches a month costs a median $4.00 per 1,000; at 1 million a month the median falls to $1.00. **SerpCrypto**, which publishes this index, charges $0.75/1K searches on its smallest pack. ## The index in five numbers - **$2.00** median entry price per 1,000 searches (19 providers) - **$4.00** median per 1,000 at 10,000 searches a month - **$1.25** median per 1,000 at 100,000 searches a month (18 providers with a self-serve option) - **$1.00** median per 1,000 at 1 million searches a month (15 providers with a self-serve option) - **42×** gap between the highest and lowest entry price ## How was the index built? We read the public pricing page of each provider on 2026-10-22 and kept only figures shown there. A "search" is one Google web results page: page 1, about 10 organic results, desktop, with no paid add-on. - **Entry price:** the least expensive paid self-serve plan or pack, divided by the searches it includes. Credit multipliers are applied, for example 25 credits per Google search at ScraperAPI. - **Price at a monthly volume:** the least expensive self-serve plan, pack or pay-as-you-go rate that covers that many searches every month. Prepaid packs are spread over the months they stay valid: six months for Serper, unlimited for Serply and DataForSEO. - **High-volume price:** the biggest self-serve plan listed, not "contact sales" offers. - **Medians** use one line per provider; DataForSEO counts once, with its standard queue, which returns results after a short delay rather than at once. Free tiers are listed but not counted in prices. AI search APIs and the official Google and Bing APIs are in a separate table, because they do not return the same thing. SerpCrypto is shown for reference and is not counted in the medians. ## Price per 1,000 Google searches, by monthly volume Each cell is the effective price per 1,000 searches when you buy that many searches every month. "—" means no self-serve option at that volume. | Provider | Entry price | 1K a month | 10K a month | 100K a month | 1M a month | Largest plan | | --- | --- | --- | --- | --- | --- | --- | | SerpCrypto (prepaid, never expires) | $0.75/1K searches | $0.75/1K searches | $0.75/1K searches | $0.56/1K searches | $0.38/1K searches | $0.30/1K searches | | SerpApi | $25.00 | $25.00 | $15.00 | $7.25 | $3.75 | $1.96 | | Serper | $1.00 | $8.33 | $1.00 | $0.75 | $0.50 | $0.30 | | SearchApi | $4.00 | $40.00 | $4.00 | $2.50 | $1.50 | $1.00 | | DataForSEO (Standard queue) | $0.60 | $0.60 | $0.60 | $0.60 | $0.60 | $0.60 | | DataForSEO (Live) | $2.00 | $2.00 | $2.00 | $2.00 | $2.00 | $2.00 | | Scrapingdog | $1.00 | $40.00 | $4.00 | $0.90 | $0.35 | $0.136 | | Zenserp | $2.00 | $49.99 | $5.00 | $1.50 | $0.90 | $0.90 | | ValueSERP | $2.50 | $2.50 | $2.50 | $2.40 | $1.00 | $1.00 | | Scale SERP | $23.00 | $23.00 | $6.60 | $5.97 | $1.70 | $1.00 | | SerpWow | $12.50 | $125.00 | $12.50 | $9.60 | $2.40 | $1.44 | | Bright Data SERP API | $1.50 | $1.50 | $1.50 | $1.50 | $1.31 | $1.31 | | Oxylabs Web Scraper API | $1.00 | $30.00 | $3.00 | $1.00 | — | $0.70 | | Decodo Web Data API | $0.75 | $19.00 | $1.90 | $0.99 | $0.50 | $0.40 | | ScraperAPI | $12.25 | $49.00 | $14.90 | $2.99 | — | $2.30 | | ScrapingBee | $2.53 | $19.00 | $4.90 | $0.99 | $1.00 | $0.48 | | HasData | $2.45 | $49.00 | $4.90 | $0.99 | — | $0.69 | | Scrapeless | $1.00 | $1.00 | $1.00 | $1.00 | $1.00 | $0.80 | | serpstack | $6.00 | $29.99 | $10.00 | $10.00 | $15.40 | $4.00 | | SerpHouse | $0.75 | $29.99 | $3.00 | — | — | $0.62 | | Serply | $2.00 | $2.00 | $1.50 | $0.75 | $0.75 | $0.75 | *Prices checked on October 22, 2026.* List prices in USD, read on 2026-10-22. Sources: each provider's pricing page, listed at the bottom. Our own prices are those of the SerpCrypto pricing page on the same date. ## Why do prices differ so much between providers? Four things explain most of the 42-fold gap at entry level. 1. **Monthly minimums.** Many providers sell plans of a fixed size. Someone who needs 1,000 searches a month pays for the whole plan, so the median at that volume is $25.00 per 1,000. 2. **Credit multipliers.** General scraping APIs charge several credits for one Google search: 5 or 10 at Scrapingdog, 10 at ScrapingBee and SerpHouse, 25 at ScraperAPI. The headline credit price hides this. 3. **Expiry.** Unused credits reset each month at most providers, are valid six months at Serper, and never expire at DataForSEO and Serply. 4. **Speed and request type.** Asynchronous queues cost less than live results, and some APIs sell a lighter request type at fewer credits than a full one. Since Google stopped honouring `num=100` in September 2025, every price here buys about 10 results. A top-100 check costs 10 searches at every provider. ## Free tiers, credit expiry, crypto and identity checks What each provider states on the pages we read. "Not mentioned" and "None stated" mean the pages did not say, not that the option does not exist. | Provider | Free tier | Credit expiry | Crypto payment | Identity checks | | --- | --- | --- | --- | --- | | SerpCrypto | None | Never | BTC, USDT, USDC, ETH, SOL, LTC | None: no ID, no selfie, no documents | | SerpApi | 250 searches a month | Monthly reset | Not mentioned | None stated | | Serper | 2,500 queries, no card | 6 months after purchase | Not mentioned | None stated | | SearchApi | 100 requests | Not stated | Not mentioned | None stated | | DataForSEO | $1 trial balance | Never | Bitcoin on request via support | None stated | | Scrapingdog | 100 credits | Lite: end of cycle; Standard and up roll over | Not mentioned | None stated | | Zenserp | 50 searches a month | Not stated | Not mentioned | None stated | | ValueSERP | Trial, size not stated | Not stated | Not mentioned | None stated | | Scale SERP | 125 searches a month | Not stated | Not mentioned | None stated | | SerpWow | Trial, size not stated | Not stated | Not mentioned | None stated | | Bright Data SERP API | 5,000 requests a month | Not stated | No | Not stated for the SERP API | | Oxylabs Web Scraper API | $5 welcome credit | Not stated | No | KYC questionnaire for every customer | | Decodo Web Data API | $5 credit | Not stated | Yes, self-serve (coins not stated) | Email verification | | ScraperAPI | 5,000 credits for 7 days | Reset at renewal | Not mentioned | None stated | | ScrapingBee | 1,000 credits | No rollover | No | None stated | | HasData | 1,000 credits a month | End of period | Not mentioned | None stated | | Scrapeless | Trial, size not stated | Unused monthly minimum not kept | Yes (coins not stated) | None for basic packages | | serpstack | 100 a month, non-commercial | Not stated | Not mentioned | None stated | | SerpHouse | 400 searches | No carry-over | Not mentioned | No | | Serply | 2,500 requests in the first 30 days | Purchased credits never expire | Not mentioned | None stated | *Prices checked on October 22, 2026.* ## What are the main findings? - **Small users pay the most.** The median cost per 1,000 searches is $25.00 at 1,000 searches a month and $1.00 at 1 million. - **Live results cost more than queued ones.** DataForSEO lists $0.60 per 1,000 on its standard queue and $2.00 for live results. - **The volume leaders differ by size.** At 10,000 searches a month, Serper and Scrapeless list $1.00; at 100,000, Serper and Serply list $0.75; at 1 million, Scrapingdog's 5-credit Light search lists $0.35. - **Crypto is rare.** Decodo Web Data API and Scrapeless state self-serve crypto payment; DataForSEO takes Bitcoin on request; Bright Data and Oxylabs say they do not accept crypto. - **Identity checks are rare too.** Oxylabs asks every customer to answer a KYC questionnaire; most others state nothing beyond an email and a card. - **Credits that never expire are the exception:** DataForSEO and Serply. ## AI search APIs and official engine APIs (not in the medians) These return different data: AI search APIs mostly answer from their own index, and the official APIs are closed or limited. Listed for buyers who compare them anyway. | API | Price per 1,000 | Note | | --- | --- | --- | | Brave Search API | $5.00 | Own index; card required "to prevent abuse"; $5 monthly credit | | Tavily | $7.50 entry, $5.00 at 100,000 | Own retrieval; monthly reset; per-call USDC via x402 | | Exa | $4.00 lite, $7.00 auto | Own index; per-request USDC via x402 | | Perplexity Search API | $5.00 | Own index | | Linkup | $5.00 standard | Per-request USDC via x402, no account | | You.com Web Search | $5.00 | Per-call USDC via x402 | | Firecrawl search | $7.60 entry, $1.50 at the largest plan | 2 credits per search | | Google Custom Search JSON API | $5.00 | Closed to new customers; ends 1 January 2027 | | Grounding with Bing Search | $14.00 per 1,000 transactions | Returns grounding for an LLM answer, not a results list | *Prices checked on October 22, 2026.* ## Can I reuse this data? Yes. The tables on this page are published under the Creative Commons Attribution 4.0 licence (CC BY 4.0): you may copy and adapt them, with a link to this page and the date 2026-10-22. Prices change often; we read them again at least once a month and update the date at the top. If you find a wrong figure, write to support@serpcrypto.com with the page that shows the right one. We correct the table and note the change in the [changelog](https://serpcrypto.com/changelog/). ## Price index: common questions ### Which SERP API costs least per 1,000 searches? It depends on volume and speed. On 2026-10-22, DataForSEO's standard queue listed $0.60 for delayed results. For live results, Serper and Scrapeless listed $1.00 at 10,000 searches a month, Serper and Serply $0.75 at 100,000, and Scrapingdog's 5-credit Light search $0.35 at 1 million. ### What is the average price of a SERP API? The median entry price across 19 providers is $2.00 per 1,000 searches. At 1 million searches a month the median is $1.00 per 1,000. Means are distorted by a few expensive monthly plans, so we publish medians. ### Why is SerpApi so much more expensive in this table? SerpApi sells monthly plans from $25 for 1,000 searches, and includes engines, legal coverage and hourly throughput guarantees that many cheaper APIs do not. Its price per 1,000 falls to $3.75 at 1 million searches a month. ### Does SerpCrypto count itself in the medians? No. SerpCrypto publishes this index and appears in the first row for reference only. The medians use the other providers. ### How often is the index updated? At least once a month, and whenever a listed provider changes its public prices. The date at the top shows the last full check. ## Sources 1. [SerpApi: pricing](https://serpapi.com/pricing) — SerpApi (read 2026-10-22) 2. [Serper: pricing](https://serper.dev/) — Serper (read 2026-10-22) 3. [SearchApi: pricing](https://www.searchapi.io/pricing) — SearchApi (read 2026-10-22) 4. [DataForSEO: pricing](https://dataforseo.com/apis/serp-api/pricing) — DataForSEO (read 2026-10-22) 5. [Scrapingdog: pricing](https://www.scrapingdog.com/pricing/) — Scrapingdog (read 2026-10-22) 6. [Zenserp: pricing](https://zenserp.com/pricing-plans/) — Zenserp (read 2026-10-22) 7. [ValueSERP: pricing](https://trajectdata.com/pricing/value-serp-api) — ValueSERP (read 2026-10-22) 8. [Scale SERP: pricing](https://trajectdata.com/pricing/scaleserp-api) — Scale SERP (read 2026-10-22) 9. [SerpWow: pricing](https://trajectdata.com/pricing/serpwow-api) — SerpWow (read 2026-10-22) 10. [Bright Data SERP API: pricing](https://www.bright.cn/pricing/serp) — Bright Data (read 2026-10-22) 11. [Oxylabs Web Scraper API: pricing](https://oxylabs.io/pricing/web-scraper-api) — Oxylabs (read 2026-10-22) 12. [Decodo Web Data API: pricing](https://decodo.com/pricing) — Decodo (read 2026-10-22) 13. [ScraperAPI: pricing](https://www.scraperapi.com/pricing/) — ScraperAPI (read 2026-10-22) 14. [ScrapingBee: pricing](https://www.scrapingbee.com/documentation/google-api/) — ScrapingBee (read 2026-10-22) 15. [HasData: pricing](https://hasdata.com/prices) — HasData (read 2026-10-22) 16. [Scrapeless: pricing](https://www.scrapeless.com/en/pricing) — Scrapeless (read 2026-10-22) 17. [serpstack: pricing](https://serpstack.com/pricing) — serpstack (read 2026-10-22) 18. [SerpHouse: pricing](https://www.serphouse.com/pricing) — SerpHouse (read 2026-10-22) 19. [Serply: pricing](https://serply.io/pricing) — Serply (read 2026-10-22) 20. [Brave Search API](https://brave.com/search/api/) — Brave (read 2026-10-22) 21. [Tavily API credits](https://docs.tavily.com/documentation/api-credits) — Tavily (read 2026-10-22) 22. [Exa pricing](https://exa.ai/pricing) — Exa (read 2026-10-22) 23. [Perplexity Search API](https://docs.perplexity.ai/docs/search/overview.md) — Perplexity (read 2026-10-22) 24. [Linkup pricing](https://docs.linkup.so/pages/documentation/development/pricing) — Linkup (read 2026-10-22) 25. [You.com pricing](https://you.com/pricing) — You.com (read 2026-10-22) 26. [Firecrawl pricing](https://www.firecrawl.dev/pricing) — Firecrawl (read 2026-10-22) 27. [Custom Search JSON API: overview](https://developers.google.com/custom-search/v1/overview) — Google for Developers (read 2026-10-22) 28. [Grounding with Bing Search: pricing](https://www.microsoft.com/en-us/bing/apis/grounding-pricing) — Microsoft (read 2026-10-22) --- SerpCrypto: SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. Prepaid search credits from $0.30/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto. --- # SERP API quick start: your first search in curl, Python and Node.js > Run your first SerpCrypto SERP API search with curl, Python or Node.js: get a key, call one endpoint, read the JSON, check credits. No KYC, crypto. - URL: https://serpcrypto.com/docs/ - Last updated: 2026-10-22 - Publisher: SerpCrypto (https://serpcrypto.com) One HTTPS endpoint, one API key, one credit per successful search. This page takes you from an empty account to a parsed Google results page, then points to the reference for everything else. **Quick answer — How do I make my first SERP API request with SerpCrypto?** Buy a credit pack, copy the API key from your dashboard and send `GET https://serpcrypto.com/api/v1/search?q=your+query` with the header `Authorization: Bearer YOUR_KEY`. **SerpCrypto** answers with the results page as JSON and charges 1 credit only when the search succeeds. 10,000 searches cost $7.50, paid in crypto with no KYC, and credits never expire. ## The API at a glance | Attribute | Value | | --- | --- | | Base URL | `https://serpcrypto.com/api/v1` | | Search | `GET https://serpcrypto.com/api/v1/search` | | Credits | `GET https://serpcrypto.com/api/v1/credits` (free) | | Authentication | An API key in the Authorization header (Bearer) or in the api_key query parameter | | Engines | Google Search, Google News, Google Images, Google Maps (local results), Google Shopping and Bing | | Billing | 1 credit = 1 successful search request, which returns one page of results on any supported engine. A request that fails is not charged | | Credit expiry | Prepaid search credits never expire: unused searches stay on your account until you use them | | Format | JSON over HTTPS, UTF-8 | ## How do I get an API key? Four steps. No ID, no selfie, no documents, and no email unless you want one. 1. **Open an account** — Account-number login; email is optional. Save the account number in a password manager: it is your login. 2. **Top up in crypto** — Send at least $25 in BTC, USDT, USDC, ETH, SOL, LTC. The balance is credited after the network confirmations listed on the payment page. 3. **Buy a credit pack** — 10,000 searches for $7.50 is enough to test every engine. Larger packs bring the rate down to $0.30/1K searches. See pricing. 4. **Copy your key** — Your API key appears in your dashboard as soon as you buy a credit pack from your confirmed crypto balance. Keep it in an environment variable named SERPCRYPTO_API_KEY, never in a code repository. ## How do I run my first search with curl? Export the key once, then send one GET request. The `-G` flag with `--data-urlencode` builds the query string and encodes spaces and accents for you. The answer is a single JSON document, and `jq` keeps it readable in a terminal. The example asks Google, the default engine, for US results in English. Change `q`, `gl` or `hl` and run it again: each successful run uses one credit. *curl + jq: the first three organic results* ```bash export SERPCRYPTO_API_KEY="paste-your-key-here" curl -sG "https://serpcrypto.com/api/v1/search" \ -H "Authorization: Bearer $SERPCRYPTO_API_KEY" \ --data-urlencode "q=best espresso grinder" \ -d gl=us -d hl=en | jq '.organic_results[:3]' ``` ## How do I run the same search in Python? Use any HTTP client. With `requests`, pass the parameters as a dictionary and the key in the header. `raise_for_status()` turns an error answer, such as `401` for a wrong key, into an exception you can catch. *Python 3.9+ with requests* ```python import os import requests API_URL = "https://serpcrypto.com/api/v1/search" API_KEY = os.environ["SERPCRYPTO_API_KEY"] response = requests.get( API_URL, params={"q": "best espresso grinder", "gl": "us", "hl": "en"}, headers={"Authorization": f"Bearer {API_KEY}"}, timeout=60, ) response.raise_for_status() data = response.json() print("credits used:", data["search_metadata"]["credits_used"]) for result in data.get("organic_results", [])[:3]: print(result["position"], result["title"], result["link"]) ``` ## And in Node.js? Node.js 18 and later ship `fetch`, so no package is needed. Save the file with the `.mjs` extension to use `await` at the top level, then run `node first-search.mjs`. *Node.js 18+ with the built-in fetch* ```javascript // first-search.mjs (Node.js 18 or later: fetch and top-level await are built in) const API_URL = "https://serpcrypto.com/api/v1/search"; const params = new URLSearchParams({ q: "best espresso grinder", gl: "us", hl: "en" }); const response = await fetch(`${API_URL}?${params}`, { headers: { Authorization: `Bearer ${process.env.SERPCRYPTO_API_KEY}` }, }); if (!response.ok) { throw new Error(`HTTP ${response.status}: ${await response.text()}`); } const data = await response.json(); console.log("credits used:", data.search_metadata.credits_used); for (const result of (data.organic_results ?? []).slice(0, 3)) { console.log(result.position, result.title, result.link); } ``` ## How do I check how many credits I have left? Call `GET https://serpcrypto.com/api/v1/credits` with the same key. It is free and returns three numbers: credits left, credits used today and how many searches your account may run at the same time. Each account can run a set number of searches at the same time; the limit is shown in your dashboard and raised on request. *Read your credits* ```bash curl -s "https://serpcrypto.com/api/v1/credits" -H "Authorization: Bearer $SERPCRYPTO_API_KEY" ``` *Answer (example values)* ```json { "credits_remaining": 84210, "credits_used_today": 1290, "concurrency_limit": 10 } ``` ## What comes back from a search? Every answer has the same frame on all engines. `search_metadata` says whether the search succeeded and how many credits it used. `search_parameters` repeats the settings that were applied, defaults included. Then come one array or object per block of the results page, such as `organic_results` or `people_also_ask`. A block is present only when the page shows it, so read it with a default value. The [response format](https://serpcrypto.com/docs/response/) page describes every block and field. *Shape of a Google answer (shortened, illustrative values)* ```json { "search_metadata": { "id": "srch_7f3a9c2e41b8", "status": "success", "engine": "google", "credits_used": 1 }, "search_parameters": { "engine": "google", "q": "best espresso grinder", "gl": "us", "hl": "en", "page": 1 }, "search_information": { "total_results": 18600000, "query_displayed": "best espresso grinder" }, "organic_results": [ { "position": 1, "title": "…", "link": "https://www.example.com/…", "snippet": "…" } ], "people_also_ask": [ { "question": "…", "snippet": "…", "link": "https://www.example.org/…" } ], "related_searches": [ { "query": "best espresso grinder under 300" } ], "pagination": { "current": 1, "next_page": 2 } } ``` ## Which parameters will I use first? The full list, with defaults and allowed values, is in the [API reference](https://serpcrypto.com/docs/search/). | Parameter | Example | What it does | | --- | --- | --- | | `q` | `best espresso grinder` | The query, as a person would type it. Required. | | `engine` | `google_news` | Which engine and vertical to search. Default `google`. | | `gl` | `de` | Country of the results, two letters. Default `us`. | | `hl` | `de` | Language of the interface. Default `en`. | | `location` | `Austin, Texas, United States` | A city-level place for local results. | | `device` | `mobile` | `desktop` (default) or `mobile` results. | | `page` | `2` | Results page, 1 to 10. Each page is one credit. | ## What does a search cost, and when is it free? 1 credit = 1 successful search request, which returns one page of results on any supported engine. A request that fails is not charged. About 10 organic results per page, as Google has served since it stopped honouring num=100 in September 2025; each extra page is one more credit. A request rejected for a bad parameter, a wrong key, an empty balance or a busy account is free as well, and so is every call to the credits endpoint. Prices run from $7.50 for 10,000 searches to $0.30/1K searches on the largest pack, as of October 2026. Prepaid search credits never expire: unused searches stay on your account until you use them. The [errors page](https://serpcrypto.com/docs/errors/) lists each error code and the right way to retry. ## Where do I go next? - [API reference: GET /search](https://serpcrypto.com/docs/search/) — every parameter, default and error - [Response format](https://serpcrypto.com/docs/response/) — each block and field of the JSON - [Locations and languages](https://serpcrypto.com/docs/locations/) — gl, hl, location, device - [Errors and retries](https://serpcrypto.com/docs/errors/) — codes, causes, backoff - [Python](https://serpcrypto.com/docs/python/) — requests, httpx async, CSV export - [Node.js](https://serpcrypto.com/docs/node/) — fetch, a concurrency pool, pagination - [curl and jq](https://serpcrypto.com/docs/curl/) — one-line recipes - [MCP server](https://serpcrypto.com/docs/mcp/) — web search in Claude Desktop and Cursor - [LangChain and LlamaIndex](https://serpcrypto.com/docs/langchain/) — a search tool for agents - [OpenAI function calling](https://serpcrypto.com/docs/openai/) — tool schema and the call loop ## Quick start questions ### Is there an official SerpCrypto SDK? No. The docs give copy-paste code for Python, Node.js and curl that calls the HTTPS endpoint directly, so there is no package to install or keep up to date. Any language with an HTTP client works the same way: one GET request, one header, one JSON answer. ### Can I put the API key in the URL? Yes. `api_key=YOUR_KEY` works as a query parameter, which helps for a quick test in a browser. In code, prefer the `Authorization: Bearer` header: URLs end up in logs, browser history and shared screenshots more often than headers do. ### Do I need KYC or a credit card? No. SerpCrypto asks for no ID, no selfie, no documents and no card. Account-number login; email is optional. You pay in BTC, USDT, USDC, ETH, SOL, LTC from a $25 minimum top-up. ### How do I test before buying a large pack? No free tier: test with the smallest pack, paid from the minimum top-up; the part added only to reach the minimum is refundable on request within 30 days, in USDT or USDC, if unspent. The smallest pack runs on the same endpoint and engines as the largest one, so a test with it tells you what production will look like. --- SerpCrypto: SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. Prepaid search credits from $0.30/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto. --- # SERP API with curl and jq: one-line recipes > One-line SerpCrypto SERP API recipes with curl and jq: top 10 links, People Also Ask, AI Overview sources, local pack and Maps, plus a PowerShell version. - URL: https://serpcrypto.com/docs/curl/ - Last updated: 2026-10-22 - Publisher: SerpCrypto (https://serpcrypto.com) The fastest way to see what the API returns is a terminal. These recipes print exactly one thing each: links, questions, sources or local businesses. They work in bash and zsh; a PowerShell version closes the page. **Quick answer — How do I query a SERP API with curl?** Run `curl -sG "https://serpcrypto.com/api/v1/search" -H "Authorization: Bearer $KEY" --data-urlencode "q=your query"` and pipe the JSON to `jq`. With **SerpCrypto**, each successful search costs 1 credit, from $0.30/1K searches, and errors are free. `-G` turns the data flags into a query string. Credits are paid in crypto, with no KYC. ## Which curl flags matter here? Set the key once per shell with `export SERPCRYPTO_API_KEY=…`. Install `jq` from your package manager if it is missing. | Flag | Why | | --- | --- | | `-s` | Silent: no progress bar mixed with the JSON. | | `-G` | Send the `-d` and `--data-urlencode` values as a query string on a GET request. | | `--data-urlencode` | Encode spaces, quotes and accents in `q` and `location`. | | `-H "Authorization: Bearer …"` | Your API key. Keep it in an environment variable. | | `-w '%{http_code}'` | Print the HTTP status, useful when a call fails. | ## How do I print the top 10 links for a keyword? Ask for the organic results and let `jq` print the position and link of each one, separated by a tab. The `[]?` form prints nothing instead of failing when a block is absent, which keeps scripts quiet on unusual pages. *Top 10 links* ```bash # Top 10 organic links, one per line curl -sG "https://serpcrypto.com/api/v1/search" \ -H "Authorization: Bearer $SERPCRYPTO_API_KEY" \ --data-urlencode "q=kubernetes cost monitoring" -d gl=us -d hl=en \ | jq -r '.organic_results[]? | "\(.position)\t\(.link)"' ``` ## How do I list the People Also Ask questions? Read `people_also_ask[].question`. The questions follow the language in `hl` and change with the country in `gl`, so run the same query in each market you write for. Content teams paste this list straight into a brief. *People Also Ask* ```bash # People Also Ask questions, ready for a content brief curl -sG "https://serpcrypto.com/api/v1/search" \ -H "Authorization: Bearer $SERPCRYPTO_API_KEY" \ --data-urlencode "q=how to brew cold brew" -d gl=gb -d hl=en \ | jq -r '.people_also_ask[]?.question' ``` ## How do I see which sources an AI Overview cites? Read `ai_overview.references`. The block exists only when Google shows an AI Overview for that query and place, so the recipe prints a clear message otherwise. Tracking these sources over time shows whether your pages are cited in Google's generated answers. *AI Overview sources* ```bash # Sources cited by the AI Overview, if the page shows one curl -sG "https://serpcrypto.com/api/v1/search" \ -H "Authorization: Bearer $SERPCRYPTO_API_KEY" \ --data-urlencode "q=what is a heat pump" -d gl=us -d hl=en \ | jq -r 'if .ai_overview then .ai_overview.references[] | "\(.index)\t\(.source)\t\(.link)" else "no AI Overview for this query" end' ``` ## How do I get the local pack and the Maps list? A web search with `location` returns the local pack shown on the results page, usually a few businesses. `engine=google_maps` returns the longer Maps list with address, rating and review count, which `@tsv` turns into rows for a spreadsheet. Both are 1 credit each. *Local pack, then Google Maps* ```bash # Local pack in a city, then the full Maps list for the same query curl -sG "https://serpcrypto.com/api/v1/search" \ -H "Authorization: Bearer $SERPCRYPTO_API_KEY" \ --data-urlencode "q=vegan restaurant" \ --data-urlencode "location=Toronto, Ontario, Canada" -d gl=ca -d hl=en \ | jq -r '.local_results[]? | "\(.position)\t\(.title)\t\(.rating // "-")"' curl -sG "https://serpcrypto.com/api/v1/search" \ -H "Authorization: Bearer $SERPCRYPTO_API_KEY" \ -d engine=google_maps --data-urlencode "q=vegan restaurant Toronto" -d gl=ca -d hl=en \ | jq -r '.local_results[]? | [.title, .address, .rating, .reviews] | @tsv' ``` ## How do I see the error when a call fails? Add `-w '\nHTTP %{http_code}\n'` to print the status after the body. An error body is small JSON with `code` and `message`; the [errors page](https://serpcrypto.com/docs/errors/) lists all six codes. The second command reads your credits, which is free. *Status code and credits* ```bash # Print the body, then the HTTP status on its own line curl -sG "https://serpcrypto.com/api/v1/search" \ -H "Authorization: Bearer $SERPCRYPTO_API_KEY" \ --data-urlencode "q=test" -w '\nHTTP %{http_code}\n' # Credits left (free call) curl -s "https://serpcrypto.com/api/v1/credits" -H "Authorization: Bearer $SERPCRYPTO_API_KEY" | jq . ``` ## What is the PowerShell equivalent on Windows? `Invoke-RestMethod` sends the request and parses the JSON into objects, so you do not need `jq`. Pass the parameters as a hashtable in `-Body`: with a GET request, PowerShell encodes them into the query string. `Format-Table` then prints positions, titles and links. *PowerShell* ```powershell # Windows PowerShell 5.1 or PowerShell 7 $headers = @{ Authorization = "Bearer $env:SERPCRYPTO_API_KEY" } $query = @{ q = "kubernetes cost monitoring"; gl = "us"; hl = "en" } $data = Invoke-RestMethod -Uri "https://serpcrypto.com/api/v1/search" -Headers $headers -Body $query -Method Get $data.organic_results | Select-Object position, title, link | Format-Table -AutoSize ``` ## When should I move from curl to code? As soon as you repeat a command with a list of keywords. A shell loop works for a few dozen searches, but retries, parallelism and storage are easier in [Python](https://serpcrypto.com/docs/python/) or [Node.js](https://serpcrypto.com/docs/node/). The requests are the same, so a recipe tested here carries over unchanged. Each search stays at 1 credit, at $0.75/1K searches on the 10,000 searches pack. ## curl questions ### Why does my query return a 400 error in curl? Usually because `q` was sent with `-d` and contains a character that needs encoding. Use `--data-urlencode "q=…"` for the query and for `location`; plain codes such as `gl=us` can stay with `-d`. ### Can I save the full JSON to a file? Yes: add `-o result.json` instead of piping to `jq`. Saving raw answers lets you re-parse them later without paying for the search again. ### Does curl on Windows work the same way? Windows 10 and 11 include `curl.exe`, and the flags are the same. Quoting differs in `cmd.exe`, so PowerShell with `Invoke-RestMethod` is usually simpler there. --- SerpCrypto: SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. Prepaid search credits from $0.30/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto. --- # SERP API errors: codes, causes, fixes and retries > SerpCrypto SERP API error codes: 400, 401, 402, 429, 502 and 504, what causes each, how to fix or retry it, and why no error ever uses a credit. - URL: https://serpcrypto.com/docs/errors/ - Last updated: 2026-10-22 - Publisher: SerpCrypto (https://serpcrypto.com) Six error codes, one rule: an error never uses a credit. This page says what triggers each code, whether to retry, and gives retry code with exponential backoff in Python and Node.js. **Quick answer — Which SERP API errors should I retry, and are failed searches charged?** **SerpCrypto** never charges an error: only a successful search uses 1 credit. Retry `429 concurrency_limit`, `502 search_failed` and `504 search_timeout` with exponential backoff. Do not retry `400`, `401` or `402`: fix the parameter, the key or the balance first. Credits cost from $0.30/1K searches and never expire; payment is in crypto, with no KYC. ## What does each error code mean? No row of this table uses a credit. 1 credit = 1 successful search request, which returns one page of results on any supported engine. A request that fails is not charged. | HTTP | `code` | Cause | What to do | Retry? | | --- | --- | --- | --- | --- | | 400 | `invalid_parameter` | `q` is missing, or a value is outside the allowed list | Fix the request; the message names the parameter | No | | 401 | `invalid_api_key` | No key, a mistyped key or a revoked key | Check the header or `api_key`; create a new key if needed | No | | 402 | `insufficient_credits` | The account has 0 credits | Buy a pack from your balance | After buying | | 429 | `concurrency_limit` | More searches at once than your account allows | Wait, lower parallelism, or ask support for a higher limit | Yes, with backoff | | 502 | `search_failed` | The engine or the collection failed for this request | Retry the same request | Yes, with backoff | | 504 | `search_timeout` | No answer in time | Retry the same request | Yes, with backoff | ## What does an error answer look like? Every error has the same small JSON body with a stable `code` and a human-readable `message`. Branch on `code` in your program and log `message` for people. The HTTP status matches the table above. *Body of a 402 answer* ```json { "error": { "code": "insufficient_credits", "message": "Your account has no search credits left." } } ``` ## Why do I get 400 invalid_parameter? A required value is missing or one value is outside the list the API accepts. The usual causes are an empty `q`, an engine name with a typo such as `google-news` instead of `google_news`, a `page` above 10, or a `device` other than `desktop` or `mobile`. The `message` names the parameter at fault. Compare your request with the [API reference](https://serpcrypto.com/docs/search/); nothing is charged, so you can correct and resend at once. ## Why do I get 401 invalid_api_key? The key did not reach the API, or it is not valid. Check that the header is exactly `Authorization: Bearer YOUR_KEY`, with a space after `Bearer`. Check also that your environment variable is set in the shell or service that runs the code, not only in your terminal. A revoked key keeps failing: create a new one in the dashboard. ## What should I do about 402 insufficient_credits? Buy a credit pack from your balance, then resend. If the balance is too low, top up in crypto first; the minimum top-up is $25. Read `credits_remaining` from `GET https://serpcrypto.com/api/v1/credits` before a large job to avoid stopping halfway: that call is free. ## How do I handle 429 concurrency_limit? Slow down rather than stop. The limit counts searches running at the same time, not searches per day. Each account can run a set number of searches at the same time; the limit is shown in your dashboard and raised on request. Read `concurrency_limit` from the credits endpoint and size your worker pool to it; the [Python](https://serpcrypto.com/docs/python/) and [Node.js](https://serpcrypto.com/docs/node/) pages show a pool that never exceeds it. ## When are 502 and 504 returned? `502 search_failed` means the search could not be completed for this request; `504 search_timeout` means it did not complete in time. Both are free and both are worth retrying. Wait a little longer before each new attempt, as the code below does, and give up after five attempts so a stuck job does not loop for ever. *Python: retry 429, 502 and 504 with exponential backoff* ```python import os import random import time import requests API_URL = "https://serpcrypto.com/api/v1/search" RETRYABLE = {429, 502, 504} def search(params, max_attempts=5): headers = {"Authorization": "Bearer " + os.environ["SERPCRYPTO_API_KEY"]} for attempt in range(1, max_attempts + 1): r = requests.get(API_URL, params=params, headers=headers, timeout=90) if r.status_code == 200: return r.json() error = r.json().get("error", {}) if r.status_code not in RETRYABLE or attempt == max_attempts: raise RuntimeError(f"{r.status_code} {error.get('code')}: {error.get('message')}") # exponential backoff with jitter: about 1 s, 2 s, 4 s, 8 s time.sleep(2 ** (attempt - 1) + random.random()) raise RuntimeError("unreachable") print(search({"q": "solar panel installers", "gl": "au", "hl": "en"})["search_metadata"]) ``` *Node.js: the same logic with fetch* ```javascript // retry.mjs — Node.js 18+ const API_URL = "https://serpcrypto.com/api/v1/search"; const RETRYABLE = new Set([429, 502, 504]); const sleep = (ms) => new Promise((resolve) => setTimeout(resolve, ms)); export async function search(params, maxAttempts = 5) { const url = `${API_URL}?${new URLSearchParams(params)}`; const headers = { Authorization: `Bearer ${process.env.SERPCRYPTO_API_KEY}` }; for (let attempt = 1; attempt <= maxAttempts; attempt++) { const res = await fetch(url, { headers }); if (res.ok) return res.json(); const body = await res.json().catch(() => ({})); const code = body.error?.code ?? "unknown"; if (!RETRYABLE.has(res.status) || attempt === maxAttempts) { throw new Error(`${res.status} ${code}: ${body.error?.message ?? ""}`); } // exponential backoff with jitter: about 1 s, 2 s, 4 s, 8 s await sleep(1000 * 2 ** (attempt - 1) + Math.random() * 1000); } throw new Error("unreachable"); } console.log((await search({ q: "solar panel installers", gl: "au", hl: "en" })).search_metadata); ``` ## Should my HTTP timeout be short or long? Long enough for a search to finish: a search runs live, so set your client timeout to 60 to 90 seconds rather than a few seconds. A client that gives up too early sends a second request while the first is still running, which wastes a concurrency slot and, if both succeed, two credits. ## Error questions ### Is a search charged if my program crashes before reading the answer? Yes, if the API completed the search and returned 200: the credit is used when the search succeeds, not when you read it. That is another reason to set a long client timeout. ### Do retries cost extra? Only the attempt that succeeds uses a credit. Failed attempts with 429, 502 or 504 are free, so a request that succeeds on its third try costs 1 credit in total. ### Where do I report an error that keeps coming back? Write to support@serpcrypto.com with the `search_metadata.id` of a successful search or the time, engine and parameters of the failing one. Email replies within 12 hours, 7 days a week, including help with API integration. --- SerpCrypto: SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. Prepaid search credits from $0.30/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto. --- # LangChain web search tool on a SERP API, plus a LlamaIndex version > A LangChain tool and a LlamaIndex FunctionTool that search Google through the SerpCrypto SERP API, with an agent that decides when to search. No KYC. - URL: https://serpcrypto.com/docs/langchain/ - Last updated: 2026-10-22 - Publisher: SerpCrypto (https://serpcrypto.com) Wrap one HTTPS call in a tool, describe it well, and let the agent decide when to search. The code below works with LangChain 1.0 or later and with LlamaIndex; no SerpCrypto package is needed. **Quick answer — How do I give a LangChain agent a web search tool?** Decorate a Python function with `@tool` from `langchain_core.tools`, make it call `https://serpcrypto.com/api/v1/search` and return short text, then pass it in `tools=[...]` when you create the agent. With **SerpCrypto**, each search the agent runs costs 1 credit, from $0.30/1K searches, and failed calls are free. Credits are paid in crypto, with no KYC. ## How do I write the search tool? A LangChain tool is a typed function with a docstring. The decorator turns the signature into the argument schema and the docstring into the description the model reads, so both deserve care. Name the arguments the way the model should think about them: a query, a country, a language. Return text, not the whole JSON. Five results with title, link and snippet give the model enough to answer and cite, at a fraction of the tokens of the raw answer. Return a short message on errors instead of raising, so the agent can explain or try again. *search_tool.py* ```python """A LangChain tool that searches Google through the SerpCrypto SERP API.""" import os import requests from langchain_core.tools import tool API_URL = "https://serpcrypto.com/api/v1/search" @tool def google_search(query: str, gl: str = "us", hl: str = "en") -> str: """Search Google for current information. Returns the top results with title, link and snippet. Args: query: the search query, as a person would type it. gl: two-letter country code of the results, for example us, gb or de. hl: language code, for example en, de or pt-BR. """ r = requests.get( API_URL, params={"q": query, "gl": gl, "hl": hl}, headers={"Authorization": "Bearer " + os.environ["SERPCRYPTO_API_KEY"]}, timeout=90, ) if r.status_code != 200: return f"Search failed with HTTP {r.status_code}." results = r.json().get("organic_results", [])[:5] return "\n\n".join(f"{x['title']}\n{x['link']}\n{x.get('snippet', '')}" for x in results) or "No results." ``` ## How do I let an agent decide when to search? Pass the tool to `create_agent`, available from LangChain 1.0. The model reads the description and calls the tool when the question needs fresh facts. Set `AGENT_MODEL` to a model string your account supports, such as `openai:` or `anthropic:` followed by a model name; any chat model with tool calling works. The system prompt matters as much as the code. Asking the agent to search only for information newer than its training data, and to cite links, keeps answers grounded and limits the number of paid searches. *agent.py* ```python """An agent that decides when to search (LangChain 1.0 or later).""" import os from langchain.agents import create_agent from search_tool import google_search agent = create_agent( model=os.environ["AGENT_MODEL"], # a provider:model string, such as one of your OpenAI or Anthropic models tools=[google_search], system_prompt="Answer with current facts. Use google_search when you need information newer than your training data, and cite the links you used.", ) result = agent.invoke({"messages": [{"role": "user", "content": "What changed in Google's search results pages this year?"}]}) print(result["messages"][-1].content) ``` ## How do I do the same with LlamaIndex? Write a plain function with type hints and a docstring, then wrap it with `FunctionTool.from_defaults`. LlamaIndex reads the name, the hints and the docstring to build the tool schema, as LangChain does. Add the tool to any LlamaIndex agent that accepts a list of tools. *LlamaIndex FunctionTool* ```python """The same search as a LlamaIndex FunctionTool.""" import os import requests from llama_index.core.tools import FunctionTool def google_search(query: str, gl: str = "us", hl: str = "en") -> str: """Search Google for current information and return the top results with title, link and snippet.""" r = requests.get( "https://serpcrypto.com/api/v1/search", params={"q": query, "gl": gl, "hl": hl}, headers={"Authorization": "Bearer " + os.environ["SERPCRYPTO_API_KEY"]}, timeout=90, ) r.raise_for_status() results = r.json().get("organic_results", [])[:5] return "\n\n".join(f"{x['title']}\n{x['link']}\n{x.get('snippet', '')}" for x in results) search_tool = FunctionTool.from_defaults(fn=google_search) # Pass search_tool in the tools list of your LlamaIndex agent. ``` ## Should the tool return JSON or text? Text, in most cases. Models read short, labelled text well, and it costs fewer tokens than nested JSON. Return JSON only when a later step of your chain parses it, for example to store rankings. Keep the links in the output. They let the model cite its sources, and they let you check afterwards which pages an answer relied on. ## How do I control search costs in an agent? Every tool call that succeeds is one credit. Three habits keep the count low: tell the model in the system prompt when not to search, cap the number of tool calls per run in your agent loop, and cache answers for repeated questions in your own code. A pack of 100,000 searches costs $56.25, and Prepaid search credits never expire: unused searches stay on your account until you use them, so a quiet month does not waste anything. ## LangChain questions ### Is this an official LangChain integration? No. It is a small custom tool built on LangChain's public `@tool` decorator. Because it is your code, you can change the output format, add parameters such as `location` or `engine`, or add a second tool for Google News. ### Can the tool search a specific city? Yes. Add a `location` argument and pass it in `params`, written `City, Region, Country`. See [locations](https://serpcrypto.com/docs/locations/) for the format and examples. ### Does the agent pay for failed searches? No. 1 credit = 1 successful search request, which returns one page of results on any supported engine. A request that fails is not charged. A tool call that hits a timeout or a temporary error uses no credit. ### Which models can use this tool? Any chat model that supports tool calling in LangChain or LlamaIndex. The tool itself only sends an HTTPS request, so it does not depend on the model provider. --- SerpCrypto: SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. Prepaid search credits from $0.30/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto. --- # Locations and languages: gl, hl, location, google_domain and device > Target Google and Bing results by country, language, city and device with the SerpCrypto SERP API: gl, hl, location and google_domain for 25 markets. - URL: https://serpcrypto.com/docs/locations/ - Last updated: 2026-10-22 - Publisher: SerpCrypto (https://serpcrypto.com) Four parameters decide whose results you get: the country, the interface language, the city and the device. This page explains each one and lists the codes for 25 common markets. **Quick answer — How do I get Google results for a specific country and city through a SERP API?** With **SerpCrypto**, set `gl` to the country code, `hl` to the language and `location` to a city written `City, Region, Country`, such as `Austin, Texas, United States`. Add `device=mobile` for phone results. Each combination is one search at 1 credit, from $0.30/1K searches, and the answer echoes the settings it applied. Paid in crypto, with no KYC. ## What does each location parameter do? Country (gl), language (hl), Google domain and a city-level location string, on desktop or mobile. | Parameter | Default | Effect | | --- | --- | --- | | `gl` | `us` | Country of the results. Two letters, ISO 3166-1, lower case. Use `gb` for the United Kingdom. | | `hl` | `en` | Language of the interface: labels, People Also Ask and some snippets follow it. | | `location` | none | A precise place inside the country, for local packs and city-level rankings. | | `google_domain` | domain of `gl` | Which Google domain runs the search, such as `google.de`. Ignored for Bing. | | `device` | `desktop` | `desktop` or `mobile`: layouts and local blocks differ between the two. | ## What is the difference between gl, hl and location? `gl` says which country's results you want. `hl` says which language the person reads, which changes the interface and often the questions and snippets. `location` narrows the country down to a city or region, which matters as soon as a query has a local intent. They combine freely. A French speaker in Switzerland is `gl=ch`, `hl=fr`. An English-language search from Berlin is `gl=de`, `hl=en`, `location=Berlin, Berlin, Germany`. Keep `gl` and the country of `location` the same: a city in one country with `gl` of another gives results that no real visitor sees. ## How do I target a city? Write the place in English, from the smallest unit to the largest: `City, Region, Country`. Examples: `Austin, Texas, United States`, `Manchester, England, United Kingdom`, `Munich, Bavaria, Germany`, `Lyon, Auvergne-Rhone-Alpes, France`. Use the name of the city, not a street or a postcode. Check `search_parameters.location` in the answer: it shows the place that was applied. A value that cannot be used returns `400 invalid_parameter` and costs nothing, so you can test a list of places cheaply. *curl + jq: the local pack for one query in two cities* ```bash # The same query from two cities: compare the local packs for city in "Austin, Texas, United States" "Denver, Colorado, United States"; do curl -sG "https://serpcrypto.com/api/v1/search" \ -H "Authorization: Bearer $SERPCRYPTO_API_KEY" \ --data-urlencode "q=emergency plumber" \ --data-urlencode "location=$city" \ -d gl=us -d hl=en -d device=mobile \ | jq -r --arg city "$city" '.local_results[]? | "\($city)\t\(.position)\t\(.title)"' done ``` ## Which codes should I use for common markets? 25 frequent markets. `google_domain` defaults to the domain of `gl`, so you rarely need to set it. | Country | `gl` | Typical `hl` | Google domain | | --- | --- | --- | --- | | United States | `us` | `en` | google.com | | United Kingdom | `gb` | `en` | google.co.uk | | Canada | `ca` | `en`, `fr` | google.ca | | Australia | `au` | `en` | google.com.au | | Ireland | `ie` | `en` | google.ie | | New Zealand | `nz` | `en` | google.co.nz | | Germany | `de` | `de` | google.de | | France | `fr` | `fr` | google.fr | | Spain | `es` | `es` | google.es | | Italy | `it` | `it` | google.it | | Netherlands | `nl` | `nl` | google.nl | | Switzerland | `ch` | `de`, `fr`, `it` | google.ch | | Poland | `pl` | `pl` | google.pl | | Sweden | `se` | `sv` | google.se | | Turkey | `tr` | `tr` | google.com.tr | | Brazil | `br` | `pt-BR` | google.com.br | | Mexico | `mx` | `es` | google.com.mx | | Argentina | `ar` | `es` | google.com.ar | | India | `in` | `en`, `hi` | google.co.in | | Japan | `jp` | `ja` | google.co.jp | | South Korea | `kr` | `ko` | google.co.kr | | Indonesia | `id` | `id` | google.co.id | | Vietnam | `vn` | `vi` | google.com.vn | | Singapore | `sg` | `en` | google.com.sg | | South Africa | `za` | `en` | google.co.za | Language codes are the ones Google uses in its interface; for a bilingual country, pick the language your audience reads. ## Which Google domain should I use? Usually none: leave `google_domain` out and the API uses the domain of the `gl` country, as in the table above. Set it only to reproduce a specific setup, for example `google.com` with `gl=gb`. The domain alone does not define the country of the results. `gl` and `location` do most of the work, which is why the answer records all three in `search_parameters`. ## Do mobile and desktop results differ? Often, yes. The mobile page can order blocks differently, show more local results for nearby intents and cut snippets shorter. If your visitors mostly use phones, track `device=mobile`; for a full picture, track both and store the device with each ranking. Each device is a separate search, so tracking both doubles the credits for those keywords. At $0.56/1K searches on the 100,000 searches pack, 1,000 keywords on two devices for 30 days use 60,000 searches. ## How do I compare several countries in one script? Loop over a list of markets and keep what the API applied. The Python example sends one query to three countries with a city each and prints the domain used and the first link. *Python: one query, three markets* ```python import os import requests MARKETS = [ {"gl": "de", "hl": "de", "location": "Munich, Bavaria, Germany"}, {"gl": "fr", "hl": "fr", "location": "Lyon, Auvergne-Rhone-Alpes, France"}, {"gl": "br", "hl": "pt-BR", "location": "Curitiba, State of Parana, Brazil"}, ] for market in MARKETS: r = requests.get( "https://serpcrypto.com/api/v1/search", params={"q": "running shoes", **market}, headers={"Authorization": "Bearer " + os.environ["SERPCRYPTO_API_KEY"]}, timeout=60, ) r.raise_for_status() applied = r.json()["search_parameters"] top = r.json().get("organic_results", [])[:1] print(applied["gl"], applied.get("google_domain"), [x["link"] for x in top]) ``` ## Does Bing take the same parameters? Yes for `q`, `gl`, `hl`, `device` and `page`: use the same codes with `engine=bing`. `google_domain` does not apply to Bing and is ignored. The answer has the same frame, with `organic_results` and `related_searches`; see the [Bing page](https://serpcrypto.com/bing-search-api/) for its use cases. ## Location questions ### Can I search from a ZIP code or GPS coordinates? Not in version 1. `location` takes a city-level place name. For results around a point on a map, use `engine=google_maps` with a query that names the area, such as `dentist Brooklyn`. ### Why do I get English labels with gl=de? Because `hl` defaults to `en`. Set `hl=de` as well to get the German interface, questions and labels a person in Germany sees. ### Does a location change the price of a search? No. Every search costs 1 credit, whatever the country, city, language or device. Prepaid search credits never expire: unused searches stay on your account until you use them. ### Which code is the United Kingdom? Use `gb`, its ISO 3166-1 code. The Google domain for it is `google.co.uk`, which the API picks for you when `gl=gb`. --- SerpCrypto: SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. Prepaid search credits from $0.30/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto. --- # Google search MCP server for Claude Desktop, Cursor and other MCP clients > Add Google search to Claude Desktop, Claude Code, Cursor and other MCP clients: a 40-line Python MCP server on the SerpCrypto SERP API. No KYC, crypto. - URL: https://serpcrypto.com/docs/mcp/ - Last updated: 2026-10-22 - Publisher: SerpCrypto (https://serpcrypto.com) Give any MCP client a `google_search` tool that returns live Google results. The server is one Python file you run yourself; it calls the SerpCrypto API and returns compact text a model can read. **Quick answer — How do I add Google search to Claude Desktop or Cursor with an MCP server?** Save the 40-line Python server below, install `mcp[cli]` and `httpx`, and register it in your client's MCP settings with your **SerpCrypto** API key in `env`. The client then gets a `google_search` tool that returns live Google results. Each tool call is one search at 1 credit, from $0.30/1K searches; credits are paid in crypto with no KYC. ## The MCP server at a glance | Attribute | Value | | --- | --- | | Language | Python 3.10+ with the official `mcp` SDK (FastMCP) and `httpx` | | Transport | stdio: the client starts the script and talks to it | | Tool | `google_search(query, gl, hl, num_results)` | | Output | AI Overview text when shown, then title, link and snippet per result | | Key | Environment variable `SERPCRYPTO_API_KEY`, set in the client config | | Cost | 1 credit per tool call that succeeds; errors are free | ## What does the server do? It exposes one tool, `google_search`, over the Model Context Protocol. When the model calls it, the server sends one request to `https://serpcrypto.com/api/v1/search`, then turns the JSON into short plain text: the AI Overview if Google showed one, and the top results with title, link and snippet. Short text matters. A full results page in JSON is long, and every character sent back to the model uses context. The server keeps up to ten results and cuts the overview at 800 characters; change those limits in the code if your client has room for more. This is copy-paste code, not a published package: there is no official SerpCrypto MCP package to install. You own the file and can add tools, for example one for `engine=google_news`. *serpcrypto_mcp.py* ```python """serpcrypto_mcp.py: an MCP server that gives AI clients a Google search tool. Install once: pip install "mcp[cli]" httpx Run by your MCP client over stdio; the API key comes from the environment. """ import os import httpx from mcp.server.fastmcp import FastMCP API_URL = "https://serpcrypto.com/api/v1/search" mcp = FastMCP("serpcrypto-search") @mcp.tool() async def google_search(query: str, gl: str = "us", hl: str = "en", num_results: int = 5) -> str: """Search Google and return the top results (title, link, snippet) and the AI Overview text if any. gl is a two-letter country code (us, gb, de...), hl a language code (en, de, pt-BR...). Each call uses one SerpCrypto search credit. """ headers = {"Authorization": "Bearer " + os.environ["SERPCRYPTO_API_KEY"]} params = {"q": query, "gl": gl, "hl": hl} async with httpx.AsyncClient(timeout=90) as client: r = await client.get(API_URL, params=params, headers=headers) if r.status_code != 200: return f"Search failed with HTTP {r.status_code}: {r.text[:300]}" data = r.json() lines = [] overview = data.get("ai_overview") or {} if overview.get("text"): lines.append("AI Overview: " + overview["text"][:800]) for item in data.get("organic_results", [])[: max(1, min(num_results, 10))]: lines.append(f"{item['position']}. {item['title']}\n{item['link']}\n{item.get('snippet', '')}") return "\n\n".join(lines) or "No results." if __name__ == "__main__": mcp.run() # stdio transport ``` ## Why is the tool called google_search and not web_search? Several clients already ship a tool named `web_search` that uses their own provider. A distinct name avoids a clash and tells the model what it gets: Google results for a country and language it chooses. The docstring is the description the model reads, so it states the parameters and the cost of a call. ## How do I add it to Claude Desktop? Open the configuration file, `claude_desktop_config.json`: on macOS it is in `~/Library/Application Support/Claude/`, on Windows in `%APPDATA%\Claude\`. Add the server under `mcpServers`, with the absolute path to the script and your key in `env`, then restart Claude Desktop. If `python` is not on the path Claude Desktop sees, use the full path to the interpreter, for example the one inside your virtual environment. *claude_desktop_config.json* ```json { "mcpServers": { "serpcrypto-search": { "command": "python", "args": ["/absolute/path/to/serpcrypto_mcp.py"], "env": { "SERPCRYPTO_API_KEY": "paste-your-key-here" } } } } ``` ## How do I add it to Claude Code and Cursor? In Claude Code, register it from a terminal with `claude mcp add`; `-e` sets the environment variable and everything after `--` is the command to run. In Cursor, put the same `mcpServers` block as above in `~/.cursor/mcp.json` for every project, or in `.cursor/mcp.json` inside one project. Other MCP clients that start stdio servers take the same three things: a command, its arguments and environment variables. The [OpenClaw page](https://serpcrypto.com/docs/openclaw/) shows the commands for OpenClaw. *Claude Code* ```bash claude mcp add serpcrypto-search \ -e SERPCRYPTO_API_KEY=paste-your-key-here \ -- python /absolute/path/to/serpcrypto_mcp.py ``` ## How do I keep the API key safe? Keep it in the `env` block of the client configuration or in your shell profile, never inside `serpcrypto_mcp.py`. That way the script can be shared or committed without the key. If a key leaks, revoke it in the dashboard and create a new one: API keys you can create and revoke at any time from your dashboard. Give the MCP server its own key. Usage per key then shows how many searches the assistant makes, separate from your scripts. ## How many credits will an assistant use? One per tool call that succeeds. A chat where the model searches three times costs three credits. On the 25,000 searches pack at $18.75, that is thousands of assisted questions, and Prepaid search credits never expire: unused searches stay on your account until you use them. To cap usage, lower `num_results` (it does not change the price, but it shortens answers) and tell the model in your instructions to search only when it needs fresh facts. ## MCP questions ### Does the server work with remote MCP clients? This version uses stdio, so the client must run it on the same machine. FastMCP can also serve over HTTP; if you expose it on a network, put authentication in front of it, because anyone who reaches it spends your credits. ### Can the tool search Bing or Google News instead? Yes. Add an `engine` argument or a second tool that sends `engine=bing` or `engine=google_news` in `params`. Every engine costs the same 1 credit per search. ### Why does the client say the tool failed? Run the script by hand with the same Python to see errors. The usual causes are a missing `mcp` or `httpx` package, a wrong path in the config, or an empty `SERPCRYPTO_API_KEY`, which makes the API answer `401`. ### Do I need a credit card or KYC for the API key? No. SerpCrypto sells prepaid credits paid in BTC, USDT, USDC, ETH, SOL, LTC, with no ID, no selfie and no documents. Account-number login; email is optional. --- SerpCrypto: SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. Prepaid search credits from $0.30/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto. --- # Web search in n8n AI agents with the SerpCrypto SERP API > Add Google web search to an n8n AI Agent: HTTP Request node settings for the SerpCrypto SERP API, header auth, $fromAI and Optimize Response. No KYC, crypto. - URL: https://serpcrypto.com/docs/n8n/ - Last updated: 2026-10-22 - Publisher: SerpCrypto (https://serpcrypto.com) No custom node to install: n8n's own HTTP Request node, attached to the AI Agent as a tool, calls the SerpCrypto API and hands the model a trimmed list of results. The same node also feeds plain workflows, such as a daily rank check. **Quick answer — How do I add a web search tool to an n8n AI Agent?** Attach an HTTP Request node to the AI Agent as a tool, set it to `GET https://serpcrypto.com/api/v1/search` with a Header Auth credential (`Authorization: Bearer `), let the model fill the `q` query parameter, and turn on Optimize Response to pass only titles, links and snippets. **SerpCrypto** charges 1 credit per successful search, from $0.30/1K searches; crypto, no KYC. ## Which HTTP Request settings do I use? Option names are those of n8n's HTTP Request node documentation, read on 2026-10-22. | Setting | Value | | --- | --- | | Placement | Attached to the **AI Agent** node as a tool sub-node | | Method | `GET` | | URL | `https://serpcrypto.com/api/v1/search` | | Authentication | Generic Credential Type → **Header Auth** | | Header Auth credential | Name `Authorization`, Value `Bearer ` followed by your API key | | Send Query Parameters | On, **Using Fields Below** | | Parameter `q` | Defined by the model (star button or `$fromAI`, see below) | | Parameters `gl`, `hl` | Fixed values such as `us` and `en`, or defined by the model | | Optimize Response | On · Response type **JSON** · Field Containing Data `organic_results` · Include Fields **Selected**: `title`, `link`, `snippet` | ## Why the HTTP Request node rather than a custom node? n8n's documentation says the HTTP Request node can run as a regular node or be attached to an AI agent to be used as a tool. That covers a REST API such as SerpCrypto, which takes one GET request with a header and a few query parameters. There is no community node of ours to install or keep up to date, and the credential stays in n8n's encrypted credential store. The AI Agent node needs a chat model and at least one tool sub-node. Since n8n 1.82.0, every AI Agent works as a Tools Agent, so the agent decides on its own when to call the search tool. ## How do I store the API key? Create a **Header Auth** credential. n8n's credential documentation describes two fields, the header Name and its Value, and notes that a bearer token is header auth with the name `Authorization` and the value `Bearer `. Put `Authorization` in Name and `Bearer ` plus your SerpCrypto key in Value. n8n also lists a generic Bearer auth type; either one sends the same header. Give n8n its own API key, so you can see its usage apart from your other scripts and revoke it alone if needed. API keys you can create and revoke at any time from your dashboard. ## How does the model fill in the search query? Let the model define `q`. Next to an eligible parameter field, n8n shows a star button that hands the value to the model. You can also type the `$fromAI()` expression yourself, which takes a key, a description, a type and a default value; the description tells the model what to put there. n8n documents that `$fromAI()` works only with tools connected to the AI Agent node. Keep `gl` and `hl` fixed if your agent serves one market. Let the model set them too if users ask about different countries, and describe the expected codes, such as `gb` or `de`. *Value of the q field, as an n8n expression* ```javascript {{ $fromAI("query", "The search query, as a person would type it", "string") }} ``` ## Why turn on Optimize Response? A full results page in JSON is long, and everything the tool returns goes into the model's context. Optimize Response is a tool-only option of the HTTP Request node that cuts the data passed to the model; n8n notes it can reduce costs. With the JSON response type, set Field Containing Data to `organic_results` and select only `title`, `link` and `snippet`. The agent then reads about ten short entries per search instead of every block of the page. If your agent answers local questions, point Field Containing Data at `local_results` in a second tool instead. ## How do I use the same API in a normal workflow? Without the agent, use the HTTP Request node as a regular node with a fixed `q`, for example in a Schedule Trigger workflow that checks a keyword every morning. Then add a **Split Out** node with Field to Split Out set to `organic_results`: it turns the list into one item per result, ready for a spreadsheet or a database node. Each run is one search per keyword, so 50 keywords a day use about 1,500 credits a month. The 25,000 searches pack ($18.75) covers that more than ten times over, and Prepaid search credits never expire: unused searches stay on your account until you use them. ## How do I test the request outside n8n? Run the same request in a terminal. If it works there, any problem in n8n comes from the node settings; if it fails, the error body names the cause, as listed on the [errors page](https://serpcrypto.com/docs/errors/). Errors are never charged. *The same request with curl* ```bash # The request the HTTP Request node sends, for a test in a terminal curl -sG "https://serpcrypto.com/api/v1/search" \ -H "Authorization: Bearer $SERPCRYPTO_API_KEY" \ --data-urlencode "q=n8n self-hosted vs cloud" -d gl=us -d hl=en \ | jq '.organic_results[]? | {title, link, snippet}' ``` ## n8n questions ### Does this work on n8n Cloud and self-hosted n8n? The HTTP Request node, Header Auth credentials and the AI Agent node are core n8n features documented for both. The SerpCrypto side is a public HTTPS endpoint, so it does not matter where n8n runs. ### Can the agent search Google News or Google Maps? Yes. Add a second HTTP Request tool with the query parameter `engine` set to `google_news` or `google_maps`, and describe it so the model knows when to use it. Every engine costs the same 1 credit per search. ### What does the agent receive when a search fails? An HTTP error with a small JSON body naming the cause, such as `429 concurrency_limit`. No credit is used, so a retry later costs nothing extra. ### Do I need a credit card for the API key? No. SerpCrypto credits are paid in BTC, USDT, USDC, ETH, SOL, LTC, with no ID, no selfie and no documents. Account-number login; email is optional. ## Sources 1. [HTTP Request node](https://docs.n8n.io/integrations/builtin/core-nodes/n8n-nodes-base.httprequest/) — n8n docs (read 2026-10-22) 2. [HTTP Request credentials](https://docs.n8n.io/integrations/builtin/credentials/httprequest/) — n8n docs (read 2026-10-22) 3. [AI Agent node](https://docs.n8n.io/integrations/builtin/cluster-nodes/root-nodes/n8n-nodes-langchain.agent/) — n8n docs (read 2026-10-22) 4. [Let AI specify tool parameters with $fromAI()](https://docs.n8n.io/advanced-ai/examples/using-the-fromai-function/) — n8n docs (read 2026-10-22) 5. [Split Out node](https://docs.n8n.io/integrations/builtin/core-nodes/n8n-nodes-base.splitout/) — n8n docs (read 2026-10-22) --- SerpCrypto: SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. Prepaid search credits from $0.30/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto. --- # SERP API in Node.js: fetch, a worker pool and pagination > Call the SerpCrypto SERP API from Node.js 18+ with the built-in fetch: a typed client, a worker pool under your concurrency limit and a pagination loop. - URL: https://serpcrypto.com/docs/node/ - Last updated: 2026-10-22 - Publisher: SerpCrypto (https://serpcrypto.com) No dependency: Node.js has had `fetch` since version 18. A small client with JSDoc types for editors and TypeScript, a pool that respects your account's limit, and a pagination loop that stops when the results run out. **Quick answer — How do I use a SERP API in Node.js?** Call `fetch("https://serpcrypto.com/api/v1/search?" + new URLSearchParams({ q }))` with the header `Authorization: Bearer ` and read `await res.json()`. Node.js 18+ needs no package. **SerpCrypto** returns Google or Bing results as JSON for 1 credit per successful search, from $0.30/1K searches, and never charges errors. Paid in crypto, no KYC. ## Node.js setup at a glance | Attribute | Value | | --- | --- | | Runtime | Node.js 18 or later (built-in `fetch`, `AbortSignal.timeout`) | | Modules | ES modules: `.mjs` files or `"type": "module"` in package.json | | Key | `process.env.SERPCRYPTO_API_KEY` | | Timeout | 90 seconds with `AbortSignal.timeout(90_000)` | | Parallel searches | A pool sized to `concurrency_limit` | | Types | JSDoc typedefs, read by VS Code and `tsc --checkJs` | ## How do I write a small typed client? Put the endpoint, the header and the error handling in one module. The JSDoc `@typedef` lines describe the answer, so editors autocomplete `organic_results` and type-check your code without a build step. TypeScript users can copy the same shapes into an interface. The `search` function throws an `Error` with the HTTP status and the API's error `code` in its message. Your code can then decide to retry `429`, `502` and `504` and to stop on the others, as the [errors page](https://serpcrypto.com/docs/errors/) explains. *serp-client.mjs* ```javascript // serp-client.mjs — Node.js 18+, no dependency const API_URL = "https://serpcrypto.com/api/v1/search"; const CREDITS_URL = "https://serpcrypto.com/api/v1/credits"; const headers = { Authorization: `Bearer ${process.env.SERPCRYPTO_API_KEY}` }; /** * @typedef {Object} OrganicResult * @property {number} position * @property {string} title * @property {string} link * @property {string} [snippet] * @property {string} [date] */ /** * @typedef {Object} SearchResponse * @property {Object} search_metadata * @property {string} search_metadata.id * @property {string} search_metadata.status * @property {number} search_metadata.credits_used * @property {OrganicResult[]} [organic_results] * @property {Object} [pagination] * @property {number} pagination.current * @property {number} [pagination.next_page] */ /** * Run one search (1 credit when it succeeds). * @param {string} q * @param {Record} [params] * @returns {Promise} */ export async function search(q, params = {}) { const query = new URLSearchParams({ q, ...Object.fromEntries(Object.entries(params).map(([k, v]) => [k, String(v)])) }); const res = await fetch(`${API_URL}?${query}`, { headers, signal: AbortSignal.timeout(90_000) }); if (!res.ok) { const body = await res.json().catch(() => ({})); const err = new Error(`${res.status} ${body.error?.code ?? "error"}: ${body.error?.message ?? ""}`); err.status = res.status; throw err; } return res.json(); } /** Free: credits left, used today and the concurrency limit. */ export async function credits() { const res = await fetch(CREDITS_URL, { headers }); if (!res.ok) throw new Error(`credits: HTTP ${res.status}`); return res.json(); } ``` ## How do I run searches in parallel without hitting 429? Start as many workers as your account allows, and let each one take the next keyword from a shared list. The pool reads `concurrency_limit` from the free credits endpoint, so it never sends more searches at once than the account accepts. A failed keyword is recorded and the worker moves on. Re-run the failures afterwards: errors cost nothing, so a second pass is free until it succeeds. *batch.mjs: a pool of workers* ```javascript // batch.mjs — parallel searches, never more than the account's limit at once import { search, credits } from "./serp-client.mjs"; const keywords = ["password manager", "note taking app", "time tracking software", "invoice software"]; const { concurrency_limit: limit } = await credits(); let next = 0; const results = new Map(); async function worker() { while (next < keywords.length) { const kw = keywords[next++]; try { const data = await search(kw, { gl: "us", hl: "en" }); results.set(kw, data.organic_results?.[0]?.link ?? null); } catch (err) { results.set(kw, `failed: ${err.message}`); } } } await Promise.all(Array.from({ length: Math.min(limit, keywords.length) }, worker)); console.table(Object.fromEntries(results)); ``` ## How do I fetch several pages of results? Loop over `page` and stop as soon as `pagination.next_page` is missing. About 10 organic results per page, as Google has served since it stopped honouring num=100 in September 2025; each extra page is one more credit. A three-page check of one keyword costs three credits; a keyword with only one page of results costs one. *pages.mjs* ```javascript // pages.mjs — up to three pages, stop when results end import { search } from "./serp-client.mjs"; async function topResults(q, pages = 3, params = {}) { const rows = []; for (let page = 1; page <= pages; page++) { const data = await search(q, { ...params, page }); for (const r of data.organic_results ?? []) rows.push({ page, position: r.position, link: r.link }); if (!data.pagination?.next_page) break; // nothing more: save the credits } return rows; } console.log(await topResults("headless cms", 3, { gl: "de", hl: "en" })); ``` ## Can I call the API from a browser or an edge function? From a server or an edge function, yes: Cloudflare Workers, Vercel and Deno all provide the same `fetch`. From a browser page, no: the key would be visible to every visitor. Put a small server route in between that holds the key and forwards only the parameters you allow. The same rule applies to mobile apps. Anything shipped to a user's device can be read, so keys stay on your servers. ## How do I keep costs predictable in Node.js? Count the calls you make: one successful call is one credit, at $0.56/1K searches on the 100,000 searches pack. Check `credits_remaining` before a large batch and stop the pool when it gets close to zero rather than collecting a series of `402` answers. Prepaid search credits never expire: unused searches stay on your account until you use them, so an unused remainder is kept for the next job. ## Node.js questions ### Is there an npm package for SerpCrypto? No. The client above is about 40 lines of standard JavaScript using `fetch`, so there is nothing of ours to install. Copy it into your project and adapt it. ### Does this work with CommonJS (require)? The examples use ES modules for top-level `await`. In a CommonJS file, wrap the calls in an `async` function and use `fetch` the same way; it is global in Node.js 18+. ### How do I use it with TypeScript? Copy the JSDoc shapes into an `interface SearchResponse` and type the result of `res.json()` with it. Optional blocks such as `organic_results?` must stay optional, because a page may not show them. ### Which Node.js version do I need? Node.js 18 or later for the built-in `fetch`. `AbortSignal.timeout` is available from Node.js 17.3, so 18 covers everything on this page. --- SerpCrypto: SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. Prepaid search credits from $0.30/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto. --- # OpenAI function calling: a Google search tool in 60 lines of Python > Give an OpenAI model a Google search tool: the JSON schema, the Python tool-calling loop on the SerpCrypto SERP API, and the same schema for Claude. No KYC. - URL: https://serpcrypto.com/docs/openai/ - Last updated: 2026-10-22 - Publisher: SerpCrypto (https://serpcrypto.com) Describe a `google_search` function to the model, run the searches it asks for, send the results back. The schema is plain JSON Schema, so it carries over to Claude and other tool-calling models. **Quick answer — How do I give an OpenAI model a web search tool?** Declare a `google_search` function with a JSON Schema in `tools`, then loop: when the model returns `tool_calls`, run each search on `https://serpcrypto.com/api/v1/search` and append the results as `tool` messages until it answers. **SerpCrypto** charges 1 credit per successful search, from $0.30/1K searches, paid in crypto with no KYC. ## What does the tool schema look like? A function tool has a name, a description and parameters in JSON Schema. The model sees only these three things, so the description must say what the tool returns and when to use it. Only `query` is required; country and language are optional and default to US English in the code. *Tool definition (Chat Completions format)* ```json { "type": "function", "function": { "name": "google_search", "description": "Search Google for current information. Returns the top results with title, link and snippet. Use it for facts newer than your training data.", "parameters": { "type": "object", "properties": { "query": { "type": "string", "description": "The search query, as a person would type it." }, "gl": { "type": "string", "description": "Two-letter country code of the results, e.g. us, gb, de." }, "hl": { "type": "string", "description": "Language code, e.g. en, de, pt-BR." } }, "required": ["query"] } } } ``` ## How does the tool-calling loop work? Send the question with `tools`. If the reply contains `tool_calls`, run each one, append the assistant message and one `tool` message per call with the matching `tool_call_id`, then call the model again. When a reply has no tool call, it is the final answer. The loop below stops after three searches. A cap protects your credits if a model keeps searching, and three searches answer most factual questions. *search_agent.py* ```python """Tool-calling loop: the model asks for searches, this code runs them on SerpCrypto.""" import json import os import requests from openai import OpenAI client = OpenAI() # reads OPENAI_API_KEY MODEL = os.environ["OPENAI_MODEL"] # any OpenAI chat model with tool calling API_URL = "https://serpcrypto.com/api/v1/search" TOOLS = [ { "type": "function", "function": { "name": "google_search", "description": "Search Google for current information. Returns the top results with title, link and snippet.", "parameters": { "type": "object", "properties": { "query": {"type": "string", "description": "The search query."}, "gl": {"type": "string", "description": "Two-letter country code, e.g. us."}, "hl": {"type": "string", "description": "Language code, e.g. en."}, }, "required": ["query"], }, }, } ] def google_search(query, gl="us", hl="en"): r = requests.get( API_URL, params={"q": query, "gl": gl, "hl": hl}, headers={"Authorization": "Bearer " + os.environ["SERPCRYPTO_API_KEY"]}, timeout=90, ) if r.status_code != 200: return f"Search failed with HTTP {r.status_code}." items = r.json().get("organic_results", [])[:5] return "\n\n".join(f"{x['title']}\n{x['link']}\n{x.get('snippet', '')}" for x in items) or "No results." def ask(question, max_searches=3): messages = [{"role": "user", "content": question}] for _ in range(max_searches + 1): reply = client.chat.completions.create(model=MODEL, messages=messages, tools=TOOLS) msg = reply.choices[0].message if not msg.tool_calls: return msg.content messages.append(msg) for call in msg.tool_calls: args = json.loads(call.function.arguments) messages.append({"role": "tool", "tool_call_id": call.id, "content": google_search(**args)}) return "Stopped: search limit reached." print(ask("Which countries changed their data protection laws in 2026? Cite sources.")) ``` ## Does the same schema work with Claude and other models? Yes, with a different wrapper. Claude's tool use takes the same JSON Schema under `input_schema`, next to `name` and `description`. Most other tool-calling APIs also accept JSON Schema for parameters; only the outer field names change. The search function itself stays the same, because it only calls the SerpCrypto endpoint. *The same tool for Claude (Messages API)* ```json { "name": "google_search", "description": "Search Google for current information. Returns the top results with title, link and snippet.", "input_schema": { "type": "object", "properties": { "query": { "type": "string", "description": "The search query." }, "gl": { "type": "string", "description": "Two-letter country code, e.g. us." }, "hl": { "type": "string", "description": "Language code, e.g. en." } }, "required": ["query"] } } ``` ## What should the tool return to the model? Short text with links. Five results with title, link and snippet are usually enough for the model to answer and to cite. Returning the full JSON works but costs more tokens and gives the model more noise to read. Add the AI Overview text when present if you want the model to see Google's own summary, and keep the source links so the answer can cite pages rather than the overview. ## How do I keep search costs predictable? Count tool calls: each one that succeeds is one credit. Cap searches per question, as the loop does, and log each call with its query so you can see what the model asks for. On the 100,000 searches pack the rate is $0.56/1K searches, and Prepaid search credits never expire: unused searches stay on your account until you use them. Errors cost nothing: 1 credit = 1 successful search request, which returns one page of results on any supported engine. A request that fails is not charged. ## Function calling questions ### Can I use the Responses API instead of Chat Completions? Yes. The Responses API takes function tools with the same JSON Schema, with `name`, `description` and `parameters` at the top level of each tool. The search function and the idea of the loop do not change. ### Why not use the model provider's built-in web search? Built-in search tools choose their own sources and settings. A function on SerpCrypto gives you Google results for the country and language you pick, as JSON you can log, at a known price per search. ### Can the model choose the country? Yes. `gl` is a parameter of the tool, so the model can set it from the question, for example `gb` for a question about the United Kingdom. Keep a default in your code for questions that name no country. ### Do I need a credit card to get the search API key? No. SerpCrypto credits are paid in BTC, USDT, USDC, ETH, SOL, LTC, with no ID, no selfie and no documents. Account-number login; email is optional. --- SerpCrypto: SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. Prepaid search credits from $0.30/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto. --- # Google search results in OpenClaw through an MCP server > OpenClaw has no Google SERP web search provider built in. Add Google results through an MCP server on the SerpCrypto SERP API. No KYC, pay in crypto. - URL: https://serpcrypto.com/docs/openclaw/ - Last updated: 2026-10-22 - Publisher: SerpCrypto (https://serpcrypto.com) OpenClaw's built-in `web_search` providers do not include a Google results API. OpenClaw can, however, run saved MCP servers and expose their tools to its agent, so the SerpCrypto MCP server adds a `google_search` tool in two commands. **Quick answer — How do I add Google search results to OpenClaw?** Save the **SerpCrypto** MCP server with `openclaw mcp add`, passing your API key with `--env`, then check it with `openclaw mcp doctor --probe`. OpenClaw exposes configured MCP tools in its normal coding and messaging tool profiles, so the agent gets a `google_search` tool. Each search costs 1 credit, from $0.30/1K searches; crypto, no KYC. ## Which web search providers does OpenClaw include? OpenClaw's web tools page, read on 22 October 2026, lists these `web_search` providers: Brave, Codex Hosted Search, DeepSeek, DuckDuckGo, Exa, Firecrawl, Gemini, Grok, Kimi, MiniMax Search, Ollama Web Search, Parallel, Parallel Search (Free), Perplexity, SearXNG and Tavily. None of them is a Google results API that returns the Google page as structured data for a chosen country and language. The same page says provider IDs are validated against those declared by bundled and installed plugin manifests, so new providers come as plugins. Writing a provider plugin is a larger job than this page covers; an MCP server needs no plugin code at all. ## How do I connect the SerpCrypto MCP server? First set up the Python MCP server from the [MCP page](https://serpcrypto.com/docs/mcp/): one file, plus the `mcp` and `httpx` packages. Then save it in OpenClaw. OpenClaw keeps saved MCP server definitions under `mcp.servers` in its configuration; `openclaw mcp add` builds one from flags, with `--command`, `--arg` and `--env` for a server started over stdio. `openclaw mcp doctor --probe` confirms that the server starts and exposes its tools. Run it after any change to the path, the Python environment or the key. *Register and check the server* ```bash # 1. Save the MCP server from /docs/mcp/ in OpenClaw openclaw mcp add serpcrypto-search \ --command python \ --arg /absolute/path/to/serpcrypto_mcp.py \ --env SERPCRYPTO_API_KEY=paste-your-key-here # 2. Check that it starts and exposes its tool openclaw mcp doctor serpcrypto-search --probe ``` ## Will the OpenClaw agent see the tool? In the default profiles, yes. OpenClaw's MCP registry documentation states that embedded OpenClaw exposes configured MCP tools in the normal `coding` and `messaging` tool profiles. The `minimal` profile hides them, and `tools.deny: ["bundle-mcp"]` turns them off, so check those two settings if the tool does not appear. The tool is named `google_search`, which keeps it apart from OpenClaw's own `web_search` tool. Both can stay enabled: the descriptions tell the model which one returns Google results. ## Can I manage the server from the browser? Yes. OpenClaw's documentation describes an MCP settings page in its Control UI, where saved MCP servers can be edited and inspected. The command line and the Control UI change the same saved definitions, so use whichever you prefer. ## What will searches cost through OpenClaw? One credit per tool call that succeeds, at the rate of your pack: $0.75/1K searches on 10,000 searches, down to $0.30/1K searches on the largest pack. 1 credit = 1 successful search request, which returns one page of results on any supported engine. A request that fails is not charged. Prepaid search credits never expire: unused searches stay on your account until you use them, so an assistant that searches only now and then does not waste a monthly quota. Give OpenClaw its own API key. Usage per key then shows how many searches the assistant makes across your chat channels. ## OpenClaw questions ### Is there an official SerpCrypto plugin on ClawHub? No. This page uses OpenClaw's support for saved MCP servers with a small Python server you run yourself. Nothing is published on ClawHub. ### Can I set the country and language of the results? Yes. The `google_search` tool takes `gl` and `hl`, so the model can pick them from the conversation. See [locations](https://serpcrypto.com/docs/locations/) for the codes. ### Where does the API key live? In the server's environment, set with `--env` when you save it, not in the Python file. If a key leaks, revoke it in the SerpCrypto dashboard and save the server again with a new one. ### Do I need KYC or a card to get a key? No. Credits are paid in BTC, USDT, USDC, ETH, SOL, LTC, with no ID, no selfie and no documents. Account-number login; email is optional. ## Sources 1. [Web tools: web_search providers](https://docs.openclaw.ai/tools/web) — OpenClaw docs (read 2026-10-22) 2. [Manage saved MCP servers](https://docs.openclaw.ai/cli/mcp/registry) — OpenClaw docs (read 2026-10-22) 3. [MCP in the Control UI](https://docs.openclaw.ai/cli/mcp/control-ui) — OpenClaw docs (read 2026-10-22) 4. [Tools overview](https://docs.openclaw.ai/tools) — OpenClaw docs (read 2026-10-22) --- SerpCrypto: SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. Prepaid search credits from $0.30/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto. --- # SERP API in Python: requests, httpx and asyncio > Use the SerpCrypto SERP API in Python: a requests client with retries, async httpx with a concurrency limit, pagination and CSV export. Tested code. - URL: https://serpcrypto.com/docs/python/ - Last updated: 2026-10-22 - Publisher: SerpCrypto (https://serpcrypto.com) A reusable client with retries, a pagination loop that stops when the results end, async searches capped at your concurrency limit, and a daily CSV for rankings. Copy-paste code, syntax-checked, with no package of ours to install. **Quick answer — How do I call a SERP API from Python?** Send a GET request to `https://serpcrypto.com/api/v1/search` with `requests` or `httpx`, your key in the `Authorization: Bearer` header and the query in `params`, then read `response.json()`. **SerpCrypto** charges 1 credit per successful search, from $0.30/1K searches, and retries of failed calls are free. Pay in crypto, no KYC. The code below needs only `requests` or `httpx`. ## Python setup at a glance | Attribute | Value | | --- | --- | | Install | `pip install requests` or `pip install httpx` | | Key | Environment variable `SERPCRYPTO_API_KEY` | | Header | `Authorization: Bearer ` | | Timeout | 60 to 90 seconds: a search runs live | | Retry | 429, 502 and 504 only, with backoff; errors are free | | Parallel searches | Up to `concurrency_limit` from `GET /credits` | ## How do I set up a reusable Python client? Create one `requests.Session` with the key in its headers and a retry policy for the three temporary errors. A session reuses connections, so a loop of searches spends less time on TLS handshakes. Save the file as `serp_client.py`; the next examples import it. `Retry` from urllib3 waits longer before each new attempt and honours a `Retry-After` header if one is sent. It never retries `400`, `401` or `402`, which need a fix, not a second try. *serp_client.py* ```python """serp_client.py: a small, reusable client for the SerpCrypto SERP API (requests).""" import os import requests from requests.adapters import HTTPAdapter from urllib3.util.retry import Retry API_URL = "https://serpcrypto.com/api/v1/search" CREDITS_URL = "https://serpcrypto.com/api/v1/credits" def make_session(): session = requests.Session() session.headers["Authorization"] = "Bearer " + os.environ["SERPCRYPTO_API_KEY"] retry = Retry( total=4, backoff_factor=1.0, # waits about 1 s, 2 s, 4 s between attempts status_forcelist=(429, 502, 504), allowed_methods=("GET",), respect_retry_after_header=True, ) session.mount("https://", HTTPAdapter(max_retries=retry)) return session SESSION = make_session() def search(q, **params): """Run one search (1 credit if it succeeds) and return the parsed JSON.""" r = SESSION.get(API_URL, params={"q": q, **params}, timeout=90) r.raise_for_status() return r.json() def credits(): """Free: credits left, used today and the concurrency limit.""" r = SESSION.get(CREDITS_URL, timeout=30) r.raise_for_status() return r.json() if __name__ == "__main__": data = search("project management software", gl="us", hl="en") for item in data.get("organic_results", []): print(item["position"], item["link"]) print(credits()) ``` ## How do I collect more than 10 results? Ask for the next pages one by one. About 10 organic results per page, as Google has served since it stopped honouring num=100 in September 2025; each extra page is one more credit. Stop as soon as `pagination.next_page` is missing: there is nothing more to fetch, and you keep the credits. For most rank-tracking work, the first page or the first three pages are enough, because clicks concentrate there. Thirty results for one keyword cost three credits. *Pagination: up to three pages, early stop* ```python from serp_client import search def top_results(q, pages=3, **params): """Collect organic results over several pages: each page is 1 credit.""" rows = [] for page in range(1, pages + 1): data = search(q, page=page, **params) for item in data.get("organic_results", []): rows.append({"page": page, "position": item["position"], "title": item["title"], "link": item["link"]}) if not data.get("pagination", {}).get("next_page"): break # no further page: stop early and save credits return rows rows = top_results("crm for small business", pages=3, gl="gb", hl="en") print(len(rows), "results") ``` ## How do I run many searches in parallel? Use `httpx.AsyncClient` with an `asyncio.Semaphore` sized to your account's limit. The script reads `concurrency_limit` from the free credits endpoint, so it adapts when support raises your limit. Searches above the limit would get `429` anyway; the semaphore avoids wasting those attempts. Each task retries temporary errors with a growing pause and returns `None` after four failures, so one stubborn query never blocks the batch. *asyncio + httpx: parallel searches under the limit* ```python """Run many searches in parallel without exceeding the account's concurrency limit.""" import asyncio import os import httpx API_URL = "https://serpcrypto.com/api/v1/search" CREDITS_URL = "https://serpcrypto.com/api/v1/credits" HEADERS = {"Authorization": "Bearer " + os.environ["SERPCRYPTO_API_KEY"]} async def search_one(client, gate, q): async with gate: # at most N searches at the same time for attempt in range(4): r = await client.get(API_URL, params={"q": q, "gl": "us", "hl": "en"}) if r.status_code in (429, 502, 504): await asyncio.sleep(2 ** attempt) continue r.raise_for_status() return q, r.json() return q, None async def main(queries): async with httpx.AsyncClient(headers=HEADERS, timeout=90) as client: limits = (await client.get(CREDITS_URL)).json() gate = asyncio.Semaphore(limits["concurrency_limit"]) results = await asyncio.gather(*(search_one(client, gate, q) for q in queries)) for q, data in results: first = (data or {}).get("organic_results", [{}])[0].get("link") print(f"{q!r}: {first}") asyncio.run(main(["best vpn router", "home espresso machine", "4k monitor for coding"])) ``` ## How do I save daily rankings to a CSV file? Append one row per result, with the day, keyword, position, link and the settings the API applied. Reading `gl` and `device` from `search_parameters` rather than from your own variables means the file records what was actually used. Run the script once a day with cron or the Windows Task Scheduler. Three keywords on mobile cost three credits a day, about 90 a month. *Append today's rankings to rankings.csv* ```python import csv from datetime import date from serp_client import search KEYWORDS = ["standing desk", "ergonomic chair", "monitor arm"] FIELDS = ["day", "keyword", "position", "title", "link", "gl", "device"] with open("rankings.csv", "a", newline="", encoding="utf-8") as fh: writer = csv.DictWriter(fh, fieldnames=FIELDS) if fh.tell() == 0: writer.writeheader() for kw in KEYWORDS: data = search(kw, gl="gb", hl="en", device="mobile") applied = data["search_parameters"] for item in data.get("organic_results", []): writer.writerow({ "day": date.today().isoformat(), "keyword": kw, "position": item["position"], "title": item["title"], "link": item["link"], "gl": applied["gl"], "device": applied["device"], }) ``` ## Which Python HTTP client should I use? Use `requests` for scripts and cron jobs: it is simple and its retry adapter is mature. Use `httpx` when you need async code, such as a web app, an agent or a batch of hundreds of keywords. Both send the same request and get the same JSON; the API does not care which one you pick. Avoid very short timeouts with either library. A client that gives up after five seconds may resend a search that was still running and pay for both. ## How much will my script cost? Count searches, not results: one call is one credit when it succeeds. A job of 500 keywords, one page each, every day for 30 days, is 15,000 searches. On the 25,000 searches pack ($18.75) that leaves room to spare, and Prepaid search credits never expire: unused searches stay on your account until you use them. Check the balance from code with the `credits()` helper above; it costs nothing. ## Python questions ### Is there a SerpCrypto package on PyPI? No. The examples call the HTTPS endpoint with `requests` or `httpx`, two widely used libraries, so there is nothing of ours to install or update. The client above is about 40 lines and you own it. ### Which Python versions does this code support? Python 3.9 and later. The async example uses `asyncio.run`, available since Python 3.7, and the rest uses only the standard library plus `requests` or `httpx`. ### Can I use the API from a Jupyter notebook? Yes. Set `SERPCRYPTO_API_KEY` before starting Jupyter, or read the key with `getpass` inside the notebook so it does not end up in a saved cell. In a notebook, use `await main(...)` instead of `asyncio.run(main(...))`. ### How do I turn results into a pandas DataFrame? Pass the list of result dictionaries to `pandas.DataFrame(data["organic_results"])`. Each field becomes a column; missing optional fields such as `date` become empty cells. --- SerpCrypto: SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. Prepaid search credits from $0.30/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto. --- # SERP API response format: every block and field of the JSON > The SerpCrypto SERP API answer, field by field: search_metadata, organic_results, ai_overview, people_also_ask, local pack, ads and per-engine arrays. - URL: https://serpcrypto.com/docs/response/ - Last updated: 2026-10-22 - Publisher: SerpCrypto (https://serpcrypto.com) One frame on every engine, then one block per part of the results page. This page lists each block, its fields and their types. In the examples, IDs, counts, prices and results are illustrative. **Quick answer — What does a SerpCrypto SERP API response contain?** Each **SerpCrypto** response is one JSON object: `search_metadata` (status, engine, credits used), `search_parameters` (the settings applied), `search_information`, then one block per part of the results page, such as `organic_results`, `ai_overview`, `people_also_ask` or `local_results`. Absent parts are simply missing. One page costs 1 credit, from $0.30/1K searches; paid in crypto, no KYC. ## What does a full Google response look like? The example below is a Google search for `standing desk` in the United Kingdom, shortened to one item per array. A real answer has about ten organic results and every other block the page showed for that query, location and device. *engine=google, gl=gb: shortened, illustrative values* ```json { "search_metadata": { "id": "srch_2b81d0c6fa47", "status": "success", "engine": "google", "created_at": "2026-10-22T10:02:31Z", "credits_used": 1 }, "search_parameters": { "engine": "google", "q": "standing desk", "gl": "gb", "hl": "en", "google_domain": "google.co.uk", "device": "desktop", "page": 1 }, "search_information": { "total_results": 52400000, "query_displayed": "standing desk" }, "ai_overview": { "text": "A standing desk lets you work while standing or sitting…", "references": [ { "index": 1, "title": "How to choose a standing desk", "link": "https://www.example.co.uk/guide", "source": "Example Guide" } ] }, "ads": [ { "position": 1, "block": "top", "title": "Electric Standing Desks", "link": "https://shop.example.com/desks", "displayed_link": "shop.example.com", "description": "Free delivery on…" } ], "organic_results": [ { "position": 1, "title": "Best standing desks 2026, tested", "link": "https://www.example.co.uk/best-standing-desks", "displayed_link": "www.example.co.uk › reviews", "snippet": "We tested 14 electric and manual desks…", "date": "12 Oct 2026", "sitelinks": [ { "title": "Budget picks", "link": "https://www.example.co.uk/best-standing-desks#budget" } ] } ], "people_also_ask": [ { "question": "Is it healthy to stand all day at a desk?", "snippet": "…", "title": "Standing at work", "link": "https://www.example.org/standing" } ], "shopping_results": [ { "position": 1, "title": "Example Desk Pro 140 cm", "price": "£299.00", "source": "Example Store", "link": "https://shop.example.com/desk-pro" } ], "related_searches": [ { "query": "standing desk electric" }, { "query": "standing desk converter" } ], "pagination": { "current": 1, "next_page": 2 } } ``` ## Why are some blocks missing from my response? Because the results page did not show them. Google decides per query, place and device whether to show ads, an AI Overview, a local pack or top stories. SerpCrypto returns what the page showed and leaves the rest out, rather than sending empty placeholders. Read every block with a default value, as in the Python lines below. The same rule applies inside a block: optional fields such as `date` or `sitelinks` appear only on results that have them. *Python: read optional blocks safely* ```python data = response.json() # A block is present only when the results page shows it: read with a default. overview = data.get("ai_overview") or {} organic = data.get("organic_results", []) questions = [item["question"] for item in data.get("people_also_ask", [])] if overview: print("AI Overview sources:", [ref["link"] for ref in overview.get("references", [])]) print(len(organic), "organic results,", len(questions), "questions") ``` ## What is in search_metadata? | Field | Type | Description | | --- | --- | --- | | `id` | string | Identifier of the search. Quote it when you contact support. | | `status` | string | `success` on a 200 answer. | | `engine` | string | The engine that ran the search. | | `created_at` | string | Time of the search, ISO 8601 in UTC. | | `credits_used` | integer | Credits taken by this search: 1 on success. | ## What are search_parameters and search_information? Store `search_parameters` with your results: it records exactly which country, language, place and device a ranking came from. | Field | Type | Description | | --- | --- | --- | | `search_parameters` | object | The parameters applied, with defaults filled in: `engine`, `q`, `gl`, `hl`, `location`, `google_domain`, `device`, `page` and the others you sent. | | `search_information.total_results` | integer | The result count the engine displayed, when it displayed one. An estimate, not an exact figure. | | `search_information.query_displayed` | string | The query as the results page showed it. | ## What fields does organic_results have? Organic results are the unpaid web results. On Bing, `organic_results` has the same shape. | Field | Type | Description | | --- | --- | --- | | `position` | integer | Order of the result in this block on this page, starting at 1. | | `title` | string | The blue-link title. | | `link` | string | The URL of the result. | | `displayed_link` | string | The URL or breadcrumb as shown under the title. | | `snippet` | string | The text shown under the title. | | `date` | string | Date shown next to the snippet, as written on the page. Optional. | | `sitelinks` | array | Extra links shown under the result, each with `title` and `link`. Optional. | ## How is the AI Overview returned? `ai_overview` is present only when Google shows an AI Overview for that query and place. Many queries have none. | Field | Type | Description | | --- | --- | --- | | `ai_overview.text` | string | The text of the AI Overview as shown on the page. | | `ai_overview.references[].index` | integer | Order of the source in the overview. | | `ai_overview.references[].title` | string | Title of the cited page. | | `ai_overview.references[].link` | string | URL of the cited page. | | `ai_overview.references[].source` | string | Name of the site, as shown. | ## What do ads and people_also_ask contain? | Field | Type | Description | | --- | --- | --- | | `ads[].position` | integer | Order within its block. | | `ads[].block` | string | `top` or `bottom` of the page. | | `ads[].title`, `link`, `displayed_link` | string | Headline, landing URL and shown URL of the ad. | | `ads[].description` | string | Ad text. | | `people_also_ask[].question` | string | The question as shown. | | `people_also_ask[].snippet` | string | The answer text shown when the question is opened. | | `people_also_ask[].title`, `link` | string | Title and URL of the page the answer comes from. | ## Which other Google blocks can appear? | Block | Fields | What it is | | --- | --- | --- | | `local_results` | `position`, `title`, `address`, `rating`, `reviews`, `type` | The local pack: businesses on a map for queries with a place intent. | | `knowledge_graph` | `title`, `type`, `description`, `source.name`, `source.link` | The knowledge panel about a person, place, company or thing. | | `top_stories` | `title`, `link`, `source`, `date` | News articles shown in a carousel. | | `inline_videos` | `title`, `link`, `source`, `duration` | Video results shown inside the web results. | | `shopping_results` | `position`, `title`, `price`, `source`, `link` | Product listings shown inside the web results. | | `related_searches` | `query` | Searches suggested at the bottom of the page. | | `pagination` | `current`, `next_page` | Current page number and the next one, when there is one. | ## What do the other engines return? Each vertical keeps the same frame and adds its own array. On Google Maps, `local_results` is richer than the local pack of a web search. | `engine` | Main array | Fields | | --- | --- | --- | | `google_news` | `news_results` | `position`, `title`, `link`, `source`, `date`, `snippet`, `thumbnail` | | `google_images` | `images_results` | `position`, `title`, `original`, `original_width`, `original_height`, `thumbnail`, `source`, `link` | | `google_maps` | `local_results` | `position`, `title`, `place_id`, `address`, `phone`, `website`, `rating`, `reviews`, `type`, `hours`, `gps_coordinates.latitude`, `gps_coordinates.longitude` | | `google_shopping` | `shopping_results` | `position`, `title`, `price`, `extracted_price`, `currency`, `source`, `link`, `delivery`, `rating`, `reviews` | | `bing` | `organic_results`, `related_searches` | Same fields as the Google organic results | ## How are prices, ratings and dates typed? Values keep the form the page used whenever the page wrote them as text. `price` is a string with its currency sign, such as `£299.00`, and Google Shopping adds `extracted_price` as a number with `currency` as a code. `rating` is a number and `reviews` an integer. Dates such as `date` stay as written on the page, for example `12 Oct 2026` or `3 days ago`, so parse them in your own code if you need a timestamp. ## How do I build an overall rank across several pages? Store the page number with each result. `position` gives the order inside its block on one page, and `search_parameters.page` tells you which page that was. About 10 organic results per page, as Google has served since it stopped honouring num=100 in September 2025; each extra page is one more credit. For rank tracking, most teams keep `link`, `position`, `page`, `gl`, `location`, `device` and `created_at` per keyword and day. That is enough to chart a ranking and to explain a change by place or device. See [rank tracking](https://serpcrypto.com/use-cases/rank-tracking/) for the full method. ## Response format questions ### Is the field naming stable? Yes within version 1. New blocks or fields can be added, so ignore keys you do not know. Removing or renaming a field would come with a new version in the path, such as `/api/v2/`. ### Can I get the raw HTML of the page? No. Version 1 returns parsed JSON only. The parsed blocks cover what most tools need: results, ads, questions, the AI Overview, the local pack and the knowledge panel. ### Why does total_results differ from what I counted? It is the estimate the engine prints, not a count of the results you can page through. Pages stop long before that number; use `pagination.next_page` to know whether another page exists. ### Does a response with no organic results cost a credit? Yes, if the search succeeded: an empty page is still a result. Only errors are free. 1 credit = 1 successful search request, which returns one page of results on any supported engine. A request that fails is not charged. --- SerpCrypto: SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. Prepaid search credits from $0.30/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto. --- # SERP API reference: GET /search and GET /credits > SerpCrypto SERP API reference: GET /search parameters, engines, defaults, response blocks, error codes and GET /credits, with curl, Python and Node.js. - URL: https://serpcrypto.com/docs/search/ - Last updated: 2026-10-22 - Publisher: SerpCrypto (https://serpcrypto.com) Every parameter, default and error code of the SerpCrypto API, version 1. Base URL: `https://serpcrypto.com/api/v1`. All requests use HTTPS; all answers are JSON in UTF-8. **Quick answer — What parameters does the SerpCrypto search endpoint take?** `GET https://serpcrypto.com/api/v1/search` takes one required parameter, `q`, and nine optional ones: `engine`, `gl`, `hl`, `location`, `google_domain`, `device`, `page`, `safe` and `period`. **SerpCrypto** returns JSON and charges 1 credit per successful search on all 6 engines; errors are free. Packs start at $7.50 for 10,000 searches, paid in crypto with no KYC. ## How do I authenticate? Send your key in one of two ways. The recommended way is the header `Authorization: Bearer YOUR_KEY`. The other is the query parameter `api_key=YOUR_KEY`, handy for a quick test in a browser tab. A missing, wrong or revoked key gets `401 invalid_api_key` and costs nothing. API keys you can create and revoke at any time from your dashboard, so you can give each script or server its own key and revoke one without touching the others. ## Run a search `GET /api/v1/search` Runs one search on the chosen engine and returns the results page as JSON. A successful search uses 1 credit, whatever the engine; an error uses none. Parameters go in the query string, URL-encoded. The example asks Google for coffee roasters in Austin, Texas; IDs, counts and results are illustrative. Authentication: API key in the `Authorization: Bearer API_KEY` header. **Parameters** | Name | In | Type | Required | Description | | --- | --- | --- | --- | --- | | `q` | query | string | Yes | The search query, as a person would type it. Search operators such as quotes or `site:` are part of the query. | | `engine` | query | string | No | `google` (default), `google_news`, `google_images`, `google_maps`, `google_shopping` or `bing`. | | `gl` | query | string | No | Country of the results: ISO 3166-1 alpha-2 code in lower case, such as `us`, `de` or `br`. Default `us`. | | `hl` | query | string | No | Interface language, such as `en`, `de` or `pt-BR`. Default `en`. | | `location` | query | string | No | City-level place, written `City, Region, Country`, such as `Austin, Texas, United States`. See [locations](https://serpcrypto.com/docs/locations/). | | `google_domain` | query | string | No | Google domain to search, such as `google.de`. Default: the domain of the `gl` country. Ignored when `engine=bing`. | | `device` | query | string | No | `desktop` (default) or `mobile`. | | `page` | query | integer | No | Results page from 1 (default) to 10. Each page is a separate search and uses 1 credit. | | `safe` | query | string | No | `active` filters explicit results; `off` (default) does not. | | `period` | query | string | No | Only results from the last hour (`h`), day (`d`), week (`w`), month (`m`) or year (`y`). | | `api_key` | query | string | No | Your key, if you do not send the `Authorization` header. | **Response** `200` ```json { "search_metadata": { "id": "srch_7f3a9c2e41b8", "status": "success", "engine": "google", "created_at": "2026-10-22T09:14:05Z", "credits_used": 1 }, "search_parameters": { "engine": "google", "q": "coffee roasters", "gl": "us", "hl": "en", "location": "Austin, Texas, United States", "google_domain": "google.com", "device": "desktop", "page": 1 }, "search_information": { "total_results": 18600000, "query_displayed": "coffee roasters" }, "local_results": [ { "position": 1, "title": "Example Coffee Roasters", "address": "123 Example St, Austin, TX", "rating": 4.7, "reviews": 312, "type": "Coffee roasters" } ], "organic_results": [ { "position": 1, "title": "The 12 best coffee roasters in Austin", "link": "https://www.example.com/austin/coffee-roasters", "displayed_link": "www.example.com › austin › coffee", "snippet": "From light Nordic roasts to dark espresso blends, these local roasters…", "date": "Sep 30, 2026" }, { "position": 2, "title": "Example Roasting Co. | Fresh beans, roasted in Austin", "link": "https://www.example.org/", "displayed_link": "www.example.org", "snippet": "Small-batch coffee roasted every weekday and shipped the same day…" } ], "people_also_ask": [ { "question": "What is the oldest coffee roaster in Austin?", "snippet": "…", "title": "Austin coffee history", "link": "https://www.example.net/austin-coffee-history" } ], "related_searches": [ { "query": "coffee roasters near me" }, { "query": "coffee roasters austin wholesale" } ], "pagination": { "current": 1, "next_page": 2 } } ``` *curl* ```bash curl -sG "https://serpcrypto.com/api/v1/search" \ -H "Authorization: Bearer $SERPCRYPTO_API_KEY" \ --data-urlencode "q=coffee roasters" \ --data-urlencode "location=Austin, Texas, United States" \ -d gl=us -d hl=en -d device=desktop ``` *Python* ```python import os import requests r = requests.get( "https://serpcrypto.com/api/v1/search", params={ "q": "coffee roasters", "location": "Austin, Texas, United States", "gl": "us", "hl": "en", }, headers={"Authorization": "Bearer " + os.environ["SERPCRYPTO_API_KEY"]}, timeout=60, ) r.raise_for_status() print(r.json()["organic_results"][0]["link"]) ``` *Node.js* ```javascript const params = new URLSearchParams({ q: "coffee roasters", location: "Austin, Texas, United States", gl: "us", hl: "en", }); const res = await fetch("https://serpcrypto.com/api/v1/search?" + params, { headers: { Authorization: "Bearer " + process.env.SERPCRYPTO_API_KEY }, }); const data = await res.json(); console.log(data.organic_results?.[0]?.link); ``` ## Which engine values exist, and what does each return? The frame (`search_metadata`, `search_parameters`, `search_information`) is the same on every engine. A block appears only when the page shows it. | `engine` | Source | Main blocks in the answer | | --- | --- | --- | | `google` | Google Search | `ai_overview`, `organic_results`, `ads`, `people_also_ask`, `local_results`, `knowledge_graph`, `top_stories`, `inline_videos`, `shopping_results`, `related_searches`, `pagination` | | `google_news` | Google News | `news_results`: title, link, source, date, snippet, thumbnail | | `google_images` | Google Images | `images_results`: original URL and size, thumbnail, source page | | `google_maps` | Google Maps | `local_results`: name, address, phone, website, rating, reviews, category, hours, coordinates | | `google_shopping` | Google Shopping | `shopping_results`: title, price, extracted price, currency, merchant, delivery, rating | | `bing` | Bing | `organic_results`, `related_searches` | ## Read your credits `GET /api/v1/credits` Returns the credits left on your account, the credits used since midnight UTC and the number of searches your account may run at the same time. This call is free. Values below are examples. Authentication: API key in the `Authorization: Bearer API_KEY` header. **Response** `200` ```json { "credits_remaining": 84210, "credits_used_today": 1290, "concurrency_limit": 10 } ``` *curl* ```bash curl -s "https://serpcrypto.com/api/v1/credits" \ -H "Authorization: Bearer $SERPCRYPTO_API_KEY" ``` *Python* ```python import os import requests r = requests.get( "https://serpcrypto.com/api/v1/credits", headers={"Authorization": "Bearer " + os.environ["SERPCRYPTO_API_KEY"]}, timeout=30, ) r.raise_for_status() print(r.json()["credits_remaining"]) ``` *Node.js* ```javascript const res = await fetch("https://serpcrypto.com/api/v1/credits", { headers: { Authorization: "Bearer " + process.env.SERPCRYPTO_API_KEY }, }); const { credits_remaining } = await res.json(); console.log(credits_remaining); ``` ## Which errors can the API return? Every error has the body `{"error": {"code": "…", "message": "…"}}`. The [errors page](https://serpcrypto.com/docs/errors/) explains each one and shows retry code. | HTTP | `code` | When it happens | Charged | | --- | --- | --- | --- | | 400 | `invalid_parameter` | A required parameter is missing or a value is outside the allowed list | No | | 401 | `invalid_api_key` | The key is missing, wrong or revoked | No | | 402 | `insufficient_credits` | Your account has no credits left | No | | 429 | `concurrency_limit` | Too many searches are running at the same time on your account | No | | 502 | `search_failed` | The engine or the collection failed for this request; retry it | No | | 504 | `search_timeout` | No answer in time; retry it | No | ## What is charged, exactly? - **One successful search = one credit**, on every engine, including Google Maps and Google Shopping. - **Each results page is one search.** About 10 organic results per page, as Google has served since it stopped honouring num=100 in September 2025; each extra page is one more credit. - **Errors are free:** every answer with an HTTP status other than 200 uses no credit. - **Reading your credits is free.** Prepaid search credits never expire: unused searches stay on your account until you use them. Packs run from $7.50 for 10,000 searches to $0.30/1K searches on the largest one; see [pricing](https://serpcrypto.com/pricing/). ## How should I encode the query? Send `q` URL-encoded, like any query-string value: a space becomes `%20` or `+`, and accented letters are encoded as UTF-8. curl does this with `--data-urlencode`, Python `requests` and `httpx` do it with `params=`, and Node.js does it with `URLSearchParams`. Write the query exactly as a person would type it in the search box. Quotes, a minus sign or `site:` stay inside `q`; the engine decides how to apply them, as it would for a visitor. ## Is the API versioned? Yes. The version is part of the path: `/api/v1/`. New fields or blocks can appear in version 1 answers, so your code should ignore keys it does not know. Removing or renaming a field would come with a new version in the path, not as a change to `v1`. ## API reference questions ### Can I send a POST request with a JSON body? Not in version 1: both endpoints take GET requests with parameters in the query string. This keeps every call easy to reproduce with curl or a browser, and easy to log without the body. ### How many results does one search return? About 10 organic results per page, as Google has served since it stopped honouring num=100 in September 2025; each extra page is one more credit. To collect 30 results, request `page=1`, `page=2` and `page=3`: three searches, three credits. ### What happens when my credits run out? The next search returns `402 insufficient_credits` and uses nothing. Buy another pack from your balance on the [pricing page](https://serpcrypto.com/pricing/); your key keeps working and searches resume at once. ### Can I use the same key from several servers? Yes. A key is not tied to an IP address. What is limited is the number of searches running at the same time on the account: `concurrency_limit` in `GET /credits` gives the figure. --- SerpCrypto: SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. Prepaid search credits from $0.30/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto. --- # SERP API questions, answered > 32 answers about the SerpCrypto SERP API: price per 1,000 searches, credits that never expire, engines, AI Overviews, num=100, KYC and crypto payment. - URL: https://serpcrypto.com/faq/ - Last updated: 2026-10-22 - Publisher: SerpCrypto (https://serpcrypto.com) The questions developers, SEO teams and AI builders ask before they choose a SERP API, with short answers and links to the page that goes deeper. **Quick answer — What should I know before buying a SERP API from SerpCrypto?** **SerpCrypto** sells prepaid search credits: $7.50 for 10,000 searches, falling to $0.30/1K searches at 5 million, as of October 2026. One credit runs one successful search on any of 6 engines, failed requests are free, and credits never expire. You pay in crypto (BTC, USDT, USDC, ETH, SOL, LTC) with no KYC. ## Price and credits ### How much does the SerpCrypto SERP API cost? 10,000 searches cost $7.50 ($0.75/1K searches), 100,000 searches cost $56.25 ($0.56/1K searches) and 1,000,000 searches cost $375 ($0.38/1K searches), as of October 2026. The rate falls to $0.30/1K searches with 5,000,000 searches. All seven packs are on the [pricing page](https://serpcrypto.com/pricing/). ### What is a search credit? 1 credit = 1 successful search request, which returns one page of results on any supported engine. A request that fails is not charged. In practice, one credit equals one HTTP request to `https://serpcrypto.com/api/v1/search` that returns a results page. A request for page 3 of the same query is a separate credit. Checking your balance with `GET https://serpcrypto.com/api/v1/credits` costs nothing. ### Do SerpCrypto credits expire? No. Prepaid search credits never expire: unused searches stay on your account until you use them. That is the main difference from monthly SERP API plans, where unused searches usually reset at the end of each billing cycle. A seasonal project or an agent with irregular traffic keeps every credit it paid for. ### Is there a free tier or a free trial? No free tier. No free tier: test with the smallest pack, paid from the minimum top-up; the part added only to reach the minimum is refundable on request within 30 days, in USDT or USDC, if unspent. The smallest pack is 10,000 searches for $7.50, enough to test every engine on your real queries. Unused credits from that test stay on your account. ### Does every engine cost the same? Yes. A search on Google, Google News, Google Images, Google Maps, Google Shopping or Bing costs one credit each. You do not need to budget a different rate for maps or shopping results, and a mixed workload is priced the same way as a single-engine one. ### Am I charged when a search fails? No. Every error response is free: invalid parameters (400), a wrong key (401), no credits (402), too many parallel requests (429), a failed search (502) and a timeout (504). Only a successful results page uses a credit. The [error reference](https://serpcrypto.com/docs/errors/) explains each code and how to retry. ### Why do I need several requests to get 100 Google results? Because Google stopped honouring the `num=100` parameter in September 2025. About 10 organic results per page, as Google has served since it stopped honouring num=100 in September 2025; each extra page is one more credit. To collect the top 100 for a keyword, request `page=1` to `page=10`; tracking only the pages you need keeps the cost down. ### What is the smallest amount I can spend on SerpCrypto? The minimum top-up is $25. You can then buy any pack from your balance. If you buy 10,000 searches for $7.50, the rest of the top-up stays as balance. If part of the top-up was added only to meet the minimum and stays unspent, you can ask for it back within 30 days, in USDT or USDC. ## Engines and data ### Which search engines does SerpCrypto support? Google Search, Google News, Google Images, Google Maps (local results), Google Shopping and Bing. You pick one with the `engine` parameter: `google`, `google_news`, `google_images`, `google_maps`, `google_shopping` or `bing`. The response keeps the same envelope for all of them, so one parser handles every engine. See the [search reference](https://serpcrypto.com/docs/search/). ### Does the API return Google AI Overviews? Yes, when Google shows one on that page. The `ai_overview` object holds the text and the list of sources it cites, with title and link. If Google does not show an AI Overview for the query, location and device you asked for, the key is simply absent. See the [Google Search API page](https://serpcrypto.com/google-search-api/). ### Can I get the results a user sees in a specific city? Yes. Country (gl), language (hl), Google domain and a city-level location string, on desktop or mobile. Send `location=Austin, Texas, United States` together with `gl=us` and `hl=en`, for example. Local packs and map results depend heavily on the city, so set it for any local SEO work. The [locations guide](https://serpcrypto.com/docs/locations/) lists the formats. ### Can I get mobile results instead of desktop? Yes. Add `device=mobile`. Google and Bing lay out mobile pages differently: blocks appear in another order and some features show only on phones. Rank trackers usually record both devices for important keywords, which costs one credit per device. ### Which fields are in the JSON response? Parsed JSON: organic results, ads, People Also Ask, related searches, top stories, knowledge panel, local pack, shopping and video results, and the AI Overview text with its sources when the page shows one. Every response starts with `search_metadata` (ID, status, engine, credits used) and `search_parameters`. The [response format page](https://serpcrypto.com/docs/response/) documents every block, field by field, with an example. ### Can SerpCrypto replace the Bing Search API? For web results, yes. Microsoft retired the Bing Search APIs in August 2025. With SerpCrypto you send `engine=bing` and receive the organic results and related searches as JSON, paid per search with credits that never expire. Our [Bing Search API replacement page](https://serpcrypto.com/alternatives/bing-search-api/) compares the options. ### Is SerpCrypto an alternative to the Google Custom Search JSON API? It can be, if you need full Google results rather than a custom search engine over chosen sites. SerpCrypto returns the regular results page, with ads, People Also Ask and local results. Status, limits and prices of Google's own API are covered in [our comparison](https://serpcrypto.com/alternatives/google-custom-search-api/). ### Can I get local businesses from Google Maps? Yes, with `engine=google_maps`. Each item of `local_results` has the name, address, phone, website, rating, review count, category, opening hours and coordinates when Google shows them. Results depend on the `location` you send. See the [Google Maps results API](https://serpcrypto.com/google-maps-api/). ## Account, payment and privacy ### Do I need to verify my identity to use SerpCrypto? No. SerpCrypto, a no-KYC SERP API, asks for no ID, no selfie, no company documents and no company email. Account-number login; email is optional. You still accept the [Acceptable Use Policy](https://serpcrypto.com/acceptable-use/), and abuse leads to suspension. More on the [no-KYC page](https://serpcrypto.com/no-kyc-serp-api/). ### Which cryptocurrencies can I pay with? BTC, USDT, USDC, ETH, SOL, LTC. Tether (USDT) is accepted on Tron (TRC-20) and Ethereum (ERC-20); USD Coin (USDC) on Ethereum. There is no card, PayPal or bank payment. The [crypto payments page](https://serpcrypto.com/crypto-payments/) lists the networks and the confirmations each one needs before your balance is credited. ### How does a crypto payment get credited? You choose a top-up amount of at least $25, send the exact amount on the network shown, then click "I've sent the payment". The payment is matched by its exact amount and credited once it is confirmed on the blockchain. You then buy a credit pack from your balance. ### Can I get a refund? Yes, as a credit to your balance. Within 72 hours of purchase, a credit pack is credited back to your balance if you used less than 20% of its searches. A pack that does not work, and that support cannot fix within 24 hours, is credited back in full. Top-ups are credited to your balance and are not paid back, except where the law requires it; the part added only to reach the minimum top-up is refundable on request within 30 days, in USDT or USDC, if unspent. The [refund policy](https://serpcrypto.com/refunds/) works through three examples. ### How do I get and manage API keys? Your API key appears in your dashboard as soon as you buy a credit pack from your confirmed crypto balance. API keys you can create and revoke at any time from your dashboard. Send the key in the `Authorization: Bearer` header, or as `api_key` for a quick test. Keep it in an environment variable such as `SERPCRYPTO_API_KEY`, never in code you share. ### What does SerpCrypto log about my searches? Request metadata for 30 days. We log, for each request, the time, account, engine, query parameters, status and credits used, and keep these logs for 30 days to show your usage and settle billing disputes, then delete them. We do not sell or share your queries. We keep your account number, orders, top-up records and credit totals while your account is open, and afterwards only as long as accounting rules require. We disclose records to authorities only when the law requires it. The [privacy policy](https://serpcrypto.com/privacy/) has the full table of what is kept and for how long. ### Do you share my queries with anyone? No sale and no sharing for marketing or datasets. To run a search, our policy is to send upstream only the search parameters, never your account number, email or API key. Records are disclosed to authorities only when the law requires it, and logs older than 30 days no longer exist. ### What if I lose my SerpCrypto account number? Your account number is your login: save it somewhere safe. Add an email (optional) if you want a way to recover it. Without an email, a lost account number cannot be recovered. If you added an email, recovery goes through it. Write the account number in a password manager the day you open the account. ## Developers and AI agents ### Is there code for Python and Node.js? Yes. The docs have copy-paste code for [Python](https://serpcrypto.com/docs/python/) (requests and async httpx), [Node.js](https://serpcrypto.com/docs/node/) (native fetch) and [curl](https://serpcrypto.com/docs/curl/). There is no SDK to install: the API is one GET request with query parameters, which any HTTP client can send. ### Can Claude or another AI assistant use SerpCrypto through MCP? Yes. The [MCP guide](https://serpcrypto.com/docs/mcp/) has a short Python server that exposes a `web_search` tool to Claude Desktop, Cursor and other MCP clients. The server calls `https://serpcrypto.com/api/v1/search` with your key from an environment variable. It is example code you run yourself, not a published package. ### Does SerpCrypto work with LangChain and OpenAI function calling? Yes. The [LangChain page](https://serpcrypto.com/docs/langchain/) wraps the API as a tool for agents, and the [OpenAI page](https://serpcrypto.com/docs/openai/) gives the JSON schema of a `web_search` function and the loop that runs it. The same schema works with other tool-calling models. ### How many searches can I run in parallel? Each account can run a set number of searches at the same time; the limit is shown in your dashboard and raised on request. A request above the limit gets HTTP 429 `concurrency_limit` and is not charged. `GET https://serpcrypto.com/api/v1/credits` returns your current limit. The [Python](https://serpcrypto.com/docs/python/) and [Node.js](https://serpcrypto.com/docs/node/) guides show how to cap parallel requests. ### How fast is a SerpCrypto search? We do not quote a response time yet, because we have not measured one with a published method. Time a few requests yourself with the smallest pack, set a generous client timeout, and retry a 504 `search_timeout` after a short wait: it is not charged. A measured figure will appear on the [status page](https://serpcrypto.com/status/) with its method and date. ## Company and legal ### Who runs SerpCrypto? SerpGoat LLC runs SerpCrypto and sells its credits. Support answers at support@serpcrypto.com: Email replies within 12 hours, 7 days a week, including help with API integration. Abuse reports go to abuse@serpcrypto.com. The [about page](https://serpcrypto.com/about/) explains how results are collected and which figures we do not publish yet. ### Is using a SERP API legal? It depends on what you collect, how you use it and where you are, and some questions are being argued in US courts in 2026. Using search results for SEO, research and AI grounding is common practice. This is not legal advice: read our [overview](https://serpcrypto.com/guides/is-a-serp-api-legal/) and ask a lawyer for your case. ### What is not allowed on SerpCrypto? Searches meant to find, track or harass a private person, child sexual abuse material, spam, fraud, and any use that breaks the law where you or we operate. Reselling raw API access as your own API needs a written agreement. The [Acceptable Use Policy](https://serpcrypto.com/acceptable-use/) gives the full list. Abuse reports are reviewed by a person, and confirmed abuse leads to suspension. --- SerpCrypto: SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. Prepaid search credits from $0.30/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto. --- # SERP API glossary > SERP API glossary: SERP features, AI Overviews, gl and hl, num=100, rank tracking, LLM grounding, MCP, rate limits, KYC and stablecoins, in plain English. - URL: https://serpcrypto.com/glossary/ - Last updated: 2026-10-22 - Publisher: SerpCrypto (https://serpcrypto.com) Short definitions of the words that come up when you buy or use a SERP API, with a link to the page that shows the term in practice. **Quick answer — What does SERP API mean?** A SERP API is a web service that runs a search on an engine such as Google or Bing and returns the search engine results page (SERP) as structured data, usually JSON. **SerpCrypto** is one: $7.50 buys 10,000 searches on 6 engines, paid in crypto with no KYC, as of October 2026. The terms below cover the rest of the vocabulary. ## Search results pages **SERP** — Search engine results page: the page an engine shows for a query. It mixes organic results, ads and SERP features, and it changes with location, language, device and time. **SERP API** — An API that returns a SERP as structured data. You send the query and settings; the provider runs the search and parses the page. See [what is a SERP API](https://serpcrypto.com/guides/what-is-a-serp-api/). **Organic result** — A regular, unpaid result: a title, a link, a displayed URL and a snippet. In SerpCrypto JSON they are in `organic_results`, each with its `position`. **SERP feature** — Any block on a results page other than a plain organic result: ads, People Also Ask, local pack, knowledge panel, top stories, videos, shopping results, AI Overview. **AI Overview** — A generated summary that Google shows above the results for some queries, with links to the pages it draws on. SerpCrypto returns its text and sources in `ai_overview` when the page shows one. See the [Google Search API](https://serpcrypto.com/google-search-api/). **AI Mode** — A separate Google Search mode that answers with a generated response and lets the user ask follow-up questions. It is distinct from the AI Overview shown on the regular results page. **People Also Ask** — A box of related questions with short answers that expand on click. Content teams use it to find the questions an audience asks. Returned in `people_also_ask`. **Local pack** — The block of nearby businesses with a small map that appears for local queries. It depends strongly on the searcher's city. Returned in `local_results`; see [local SEO](https://serpcrypto.com/use-cases/local-seo/). **Knowledge panel** — The fact box about an entity (a company, a person, a place) shown beside or above the results. Returned in `knowledge_graph` with its title, type, description and source. **Featured snippet** — A highlighted answer quoted from one page, shown above the other organic results. It earns attention because it answers the query directly on the results page. **Top stories** — A carousel of recent news articles shown for newsworthy queries. Returned in `top_stories`; the [Google News API](https://serpcrypto.com/google-news-api/) gives a fuller news list. ## Search parameters **gl** — The country of the search, as a two-letter code such as `us`, `gb` or `de`. It changes which results and ads appear. See [locations and languages](https://serpcrypto.com/docs/locations/). **hl** — The interface language of the results page, such as `en`, `fr` or `pt-BR`. It affects labels, some features and which pages rank. **uule** — An encoded location value that Google search URLs can carry to set a precise place. With SerpCrypto you send a readable `location` string instead, such as `Denver, Colorado, United States`. **google_domain** — The Google site the search runs on, such as `google.com`, `google.co.uk` or `google.de`. SerpCrypto picks it from `gl` unless you set it. **num=100** — A Google URL parameter that used to return 100 results on one page. Google stopped honouring it in September 2025, so 100 results now take 10 page requests. **Pagination** — Fetching results beyond the first page with the `page` parameter. About 10 organic results per page, as Google has served since it stopped honouring num=100 in September 2025; each extra page is one more credit. ## SEO and research **Rank tracking** — Recording the position of a page for a set of keywords over time, by country, city and device. See [rank tracking with a SERP API](https://serpcrypto.com/use-cases/rank-tracking/). **Share of voice** — The part of the visible results, clicks or impressions that a brand holds across a set of keywords, compared with its competitors. **Keyword difficulty** — A score that estimates how hard it is to rank for a keyword. Each SEO tool computes it differently, often from the authority of the pages already ranking. **robots.txt** — A file at the root of a site that tells automated clients which paths they may fetch, standardised as RFC 9309. Its role for search results is discussed in [is a SERP API legal?](https://serpcrypto.com/guides/is-a-serp-api-legal/). ## APIs and AI agents **LLM grounding** — Giving a language model current, citable sources at answer time so it relies less on its training data. A search tool is the usual way. See [AI agents and LLM grounding](https://serpcrypto.com/use-cases/ai-agents/). **RAG** — Retrieval-augmented generation: the app retrieves documents or search results first, then passes them to the model with the question. Web search is one retrieval source among others. **Tool calling** — A model feature that lets the model ask the application to run a named function, such as `web_search`, with arguments it chooses. See [OpenAI function calling](https://serpcrypto.com/docs/openai/). **MCP** — Model Context Protocol: an open protocol that lets AI apps such as Claude Desktop or Cursor use external tools through small servers. See [the SerpCrypto MCP guide](https://serpcrypto.com/docs/mcp/). **API key** — A secret string that identifies your account on each request. SerpCrypto reads it from the `Authorization: Bearer` header or the `api_key` parameter. API keys you can create and revoke at any time from your dashboard. **Credit** — The unit SerpCrypto bills in. 1 credit = 1 successful search request, which returns one page of results on any supported engine. A request that fails is not charged. **Concurrency** — The number of requests running at the same time. Each account can run a set number of searches at the same time; the limit is shown in your dashboard and raised on request. **Rate limit** — A cap on how many requests a client may send in a period or at once. Exceeding it returns an error instead of a result. **HTTP 429** — The status code for too many requests. At SerpCrypto it means the account's concurrency limit was reached; the request is not charged. See [errors and retries](https://serpcrypto.com/docs/errors/). **Exponential backoff** — A retry strategy that waits longer after each failure, for example 1, 2, then 4 seconds, often with a little randomness, so retries do not pile up. **JSON** — JavaScript Object Notation, the text format of SerpCrypto responses: objects, arrays, strings and numbers that every programming language can parse. See the [response format](https://serpcrypto.com/docs/response/). ## Accounts and payments **KYC** — Know your customer: identity checks such as an ID scan or a selfie before you can buy. SerpCrypto asks for none. See the [no-KYC SERP API page](https://serpcrypto.com/no-kyc-serp-api/). **Top-up** — A payment that adds money to your account balance, from which you then buy credit packs. The minimum top-up is $25. **Stablecoin** — A cryptocurrency designed to keep a steady value against a currency, usually the US dollar. USDT and USDC are the two largest by market value. **USDT TRC-20** — Tether's dollar stablecoin on the Tron network. Fees are usually low, which makes it a common choice for top-ups. Send it only to a Tron address, never on another network. See [crypto payments](https://serpcrypto.com/crypto-payments/). --- SerpCrypto: SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. Prepaid search credits from $0.30/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto. --- # Google Images API: image results with their source pages, as JSON > Google Images API: image results as JSON with original URL, size, thumbnail and host page, by country, from $0.30/1K searches. No KYC, pay in crypto. - URL: https://serpcrypto.com/google-images-api/ - Last updated: 2026-10-22 - Publisher: SerpCrypto (https://serpcrypto.com) Run a Google Images search for any query and country, and read each result's original image URL, size, thumbnail, title and the page that hosts it. **Quick answer — Is there an API that returns Google Images results?** Yes. With **SerpCrypto**, set `engine=google_images` and the API returns the Google Images results for your query and country as JSON: title, original image URL, width and height, thumbnail, source name and the page that hosts each image. One successful search costs one credit, from $0.30/1K searches as of October 2026. No KYC; pay in crypto. ## Google Images results with SerpCrypto | Attribute | Value | | --- | --- | | Endpoint | `GET https://serpcrypto.com/api/v1/search?engine=google_images&q=…` | | Array returned | `images_results` | | Fields | `position`, `title`, `original`, `original_width`, `original_height`, `thumbnail`, `source`, `link` | | Targeting | `gl`, `hl`, `safe=active` or `off`, `period` | | What you get | Links and metadata; the images stay on their owners' servers | | Price | 1 credit per search; 10,000 searches for $7.50 | ## What does a Google Images search return? It returns an `images_results` array in the order Google Images showed the pictures. Each object describes one image and the page it comes from. | Field | Meaning | |---|---| | `position` | Rank in the image grid | | `title` | Title of the image or of its page | | `original` | URL of the full-size image file | | `original_width`, `original_height` | Size of the original in pixels | | `thumbnail` | URL of the small preview Google shows | | `source` | Name of the site that hosts the image | | `link` | URL of the page where the image appears | SerpCrypto does not copy or store the images. The API returns where they are, and you decide what to fetch. ## Who owns the images in the results? The images belong to their owners, not to Google and not to us. A search result tells you that an image exists and where it is published; it gives you no right to reuse it. Before you copy, publish or train on an image, check its licence on the host page, or ask the owner. Many photos are protected by copyright even when they are easy to find. For brand protection work, the opposite applies: the results help you find copies of your own pictures used without your permission. *Node.js: find where your product photo appears* ```javascript const API = "https://serpcrypto.com/api/v1/search"; const KEY = process.env.SERPCRYPTO_API_KEY; const params = new URLSearchParams({ engine: "google_images", q: "acme trail runner shoe", gl: "us", hl: "en" }); const res = await fetch(`${API}?${params}`, { headers: { Authorization: `Bearer ${KEY}` } }); if (!res.ok) throw new Error(`HTTP ${res.status}`); const data = await res.json(); const ownSite = "acme.example"; for (const img of data.images_results ?? []) { const host = new URL(img.link).hostname; if (!host.endsWith(ownSite)) { console.log(img.position, img.source, img.link); } } ``` ## What can I do with image results? Image results answer questions that text results cannot: - **Image SEO:** whether your product photos and diagrams rank for the queries you care about, and which competitor images rank above them. - **Brand protection:** pages outside your own sites that show your logo or product photos, so you can check whether they are allowed to. - **Merchandising research:** which product styles and photo formats dominate a category in each country. - **Content research:** which charts, diagrams and illustrations Google surfaces for a topic, as a starting point for your own original visuals. For each of these, the `link` field takes you to the page to review by hand. ## What about Google's own image search API? Google's Custom Search JSON API could run image searches, but Google closed it to new customers. Its overview page says existing customers have until 1 January 2027 to move to another solution. And since 20 January 2026, a new Programmable Search Engine can only search up to 50 chosen domains, not the entire web. Our [Google Custom Search API alternative](https://serpcrypto.com/alternatives/google-custom-search-api/) page compares the options for teams that relied on it. ## How much does the Google Images engine cost? Image searches cost one credit each, like every other engine. 10,000 searches cost $7.50 and 100,000 searches cost $56.25, as of October 2026. Credits never expire. | Plan | Quantity | Price | Price per unit | | --- | --- | --- | --- | | Test | 10,000 searches | $7.50 | $0.75/1K searches | | Starter | 25,000 searches | $18.75 | $0.75/1K searches | | Build | 100,000 searches | $56.25 | $0.56/1K searches | In every plan: Every engine at 1 credit per search · Failed requests not charged · Credits never expire · Country, language, city and device targeting **Before you pay** - Minimum top-up $25. Top up once in crypto, then buy any plan from your balance. - Unused balance stays on your account and never expires, and it pays for search credits. When a purchase costs less than the minimum top-up, the part added only to reach the minimum is refundable on request to support within 30 days of the payment, in USDT or USDC, as long as it has not been spent. The rest of the balance is not paid back, except where the law requires it. - Credited once your payment is checked on the blockchain. Typical network confirmation time: BTC ≈ 20 min · USDT TRC-20 ≈ 1 min · USDT ERC-20 ≈ 5 min · USDC ERC-20 ≈ 5 min · ETH ≈ 5 min · SOL ≈ 1 min · LTC ≈ 15 min ## Google Images API: common questions ### Can I download the images through the API? The API returns URLs, not files. You can fetch an image from its `original` URL yourself, but only where you have the right to use it. Check the licence on the host page first. ### Can I filter out explicit images? Yes. Add `safe=active` to ask Google to filter explicit results. Searches for child sexual abuse material are forbidden by our [Acceptable Use Policy](https://serpcrypto.com/acceptable-use/) and lead to suspension. ### Does the API support reverse image search? No. The Images engine takes a text query. It does not accept an uploaded picture or an image URL as the query. ### How many images does one search return? One search returns the first set of images Google shows for the query. The number is set by Google. Ask for `page=2` to get the next set; each page is one credit. ### Do I need an ID or a card? No. SerpCrypto is a no-KYC SERP API. You top up from $25 in BTC, USDT, USDC, ETH, SOL, LTC and buy a pack from your balance. ## Sources 1. [Custom Search JSON API overview](https://developers.google.com/custom-search/v1/overview) — Google for Developers (checked 2026-10-22) 2. [Programmable Search Engine: search the entire web](https://support.google.com/programmable-search/answer/12397162?hl=en) — Programmable Search Engine Help (checked 2026-10-22) --- SerpCrypto: SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. Prepaid search credits from $0.30/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto. --- # Google Maps results API: the places Google Maps lists, as JSON > Google Maps results API: business listings for any query and city as JSON, with address, rating, hours and coordinates. From $0.30/1K searches, no KYC. - URL: https://serpcrypto.com/google-maps-api/ - Last updated: 2026-10-22 - Publisher: SerpCrypto (https://serpcrypto.com) Send a search such as "coworking space" with a city, and get back the business listings Google Maps shows there: name, category, address, rating, review count, opening hours and coordinates. **Quick answer — Is there an API that returns Google Maps search results as JSON?** Yes. With **SerpCrypto**, set `engine=google_maps`, a query and a `location`, and the API returns the business listings Google Maps shows for that search as JSON: name, category, address, phone, website, rating, review count, hours and coordinates. Each successful search costs one credit, from $0.30/1K searches as of October 2026. No KYC; pay in crypto. ## Google Maps results with SerpCrypto | Attribute | Value | | --- | --- | | Endpoint | `GET https://serpcrypto.com/api/v1/search?engine=google_maps&q=…&location=…` | | Array returned | `local_results` | | Fields per place | `position`, `title`, `place_id`, `address`, `phone`, `website`, `rating`, `reviews`, `type`, `hours`, `gps_coordinates` | | Not returned | Review texts, reviewer names, photos of people | | Price | 1 credit per search; 25,000 searches for $18.75, 100,000 searches for $56.25 | | Payment | BTC, USDT, USDC, ETH, SOL, LTC; no KYC, no card | ## What does a Google Maps search return? It returns a `local_results` array with one object per business, in the order Google Maps listed them. Fields Google does not show for a business, such as a website or opening hours, are left out of that object. | Field | Meaning | |---|---| | `position` | Rank in the Maps list, starting at 1 | | `title` | Business name as displayed | | `place_id` | Google's identifier for the place, useful to match the same business over time | | `address` | Address line as displayed | | `phone`, `website` | Business contact details shown on the listing | | `rating`, `reviews` | Average star rating and number of reviews | | `type` | Main category, such as "Dental clinic" | | `hours` | Opening hours as shown | | `gps_coordinates` | `latitude` and `longitude` | The API returns the listing as Google displayed it at search time. It does not read the text of reviews. *Python: save the Maps listings for a query and a city to CSV* ```python import csv import os import requests API = "https://serpcrypto.com/api/v1/search" HEADERS = {"Authorization": "Bearer " + os.environ["SERPCRYPTO_API_KEY"]} FIELDS = ["position", "title", "type", "address", "rating", "reviews", "website"] params = { "engine": "google_maps", "q": "coworking space", "location": "Lisbon, Portugal", "gl": "pt", "hl": "en", } resp = requests.get(API, headers=HEADERS, params=params, timeout=60) resp.raise_for_status() places = resp.json().get("local_results", []) with open("coworking-lisbon.csv", "w", newline="", encoding="utf-8") as fh: writer = csv.DictWriter(fh, fieldnames=FIELDS, extrasaction="ignore") writer.writeheader() for place in places: writer.writerow(place) print(len(places), "places saved") ``` ## How is this different from Google's Places API? Google's own Places API (New) is a different product. It lets an application look up places and their details through Google's developer platform, with its own keys, quotas, terms and billing. You use it when your app needs place data as a building block, for example an address form or a store finder. The SerpCrypto Maps engine answers another question: what does a person see when they search Google Maps for this phrase in this place, and in what order? That ranking is what local SEO teams, market researchers and directory builders need to measure, and it is not what a place lookup returns. ## Should I use the map pack or the Maps engine? Use the map pack when you care about visibility in normal Google results, and the Maps engine when you need the full list of places. The map pack is the block of about three businesses inside a Google results page; it comes back in `local_results` when you call `engine=google`. The Maps engine lists many more places for the same search, with phone, website, hours and coordinates. Both cost one credit per search. Our [local SEO guide](https://serpcrypto.com/use-cases/local-seo/) shows how to track a business across several cities with either one. ## What are good uses for Maps data, and what is not allowed? Good uses work with business listings as businesses present them in public: - **Local SEO audits:** which competitors appear first for a service in each district, and with which category. - **Market research:** how many gyms, clinics or cafés Google Maps lists in an area, and how they are rated. - **Directory quality:** checking that addresses, hours and websites in your own directory match the public listing. - **Expansion planning:** comparing the density of a business type between cities. Some listings belong to sole traders, so a business name or phone number can also be personal data in some countries. Process it only for a lawful purpose and follow local data protection law. Our [Acceptable Use Policy](https://serpcrypto.com/acceptable-use/) forbids Searches meant to find, track or harass a private person, child sexual abuse material, spam, fraud, and any use that breaks the law where you or we operate. ## How much do Google Maps searches cost? The Maps engine costs one credit per successful search, the same as Google Search and Bing. 25,000 searches cost $18.75, as of October 2026, and credits never expire. | Plan | Quantity | Price | Price per unit | | --- | --- | --- | --- | | Test | 10,000 searches | $7.50 | $0.75/1K searches | | Starter | 25,000 searches | $18.75 | $0.75/1K searches | | Build | 100,000 searches | $56.25 | $0.56/1K searches | In every plan: Every engine at 1 credit per search · Failed requests not charged · Credits never expire · Country, language, city and device targeting **Before you pay** - Minimum top-up $25. Top up once in crypto, then buy any plan from your balance. - Unused balance stays on your account and never expires, and it pays for search credits. When a purchase costs less than the minimum top-up, the part added only to reach the minimum is refundable on request to support within 30 days of the payment, in USDT or USDC, as long as it has not been spent. The rest of the balance is not paid back, except where the law requires it. - Credited once your payment is checked on the blockchain. Typical network confirmation time: BTC ≈ 20 min · USDT TRC-20 ≈ 1 min · USDT ERC-20 ≈ 5 min · USDC ERC-20 ≈ 5 min · ETH ≈ 5 min · SOL ≈ 1 min · LTC ≈ 15 min ## Google Maps results API: common questions ### How many places does one Maps search return? One search returns the first list of places Google Maps shows for the query and location. Request `page=2` for the next list; each page is one credit. The number per page is set by Google, not by us. ### Can I get the text of Google reviews? No. The Maps engine returns the average rating and the number of reviews for each place, not review texts or reviewer names. Reviews are written by people and belong to their authors. ### How precise is the location? Use a city or a district name in the form `District, City, Country`. Google ranks places partly by distance, so a precise location changes the order. Very small areas may resolve to the nearest larger place. ### Is this the official Google Maps Platform? No. SerpCrypto is an independent SERP API that returns the Maps results page as JSON. Google Maps Platform products such as Places API (New) are sold by Google with their own terms. ### Do I need a company to buy credits? No. SerpCrypto is a no-KYC SERP API: no ID, no selfie, no company documents and no card. You top up from $25 in BTC, USDT, USDC, ETH, SOL, LTC. ## Sources 1. [Places API (New) overview](https://developers.google.com/maps/documentation/places/web-service/overview) — Google for Developers (checked 2026-10-22) 2. [Tips to improve your local ranking on Google](https://support.google.com/business/answer/7091) — Google Business Profile Help (checked 2026-10-22) --- SerpCrypto: SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. Prepaid search credits from $0.30/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto. --- # Google News API: headlines as JSON, by country and language > Google News API for monitoring and AI agents: headlines with source, date and link as JSON, by country and language, from $0.30/1K searches. No KYC. - URL: https://serpcrypto.com/google-news-api/ - Last updated: 2026-10-22 - Publisher: SerpCrypto (https://serpcrypto.com) Ask Google News about a brand, a topic or a person in the news, for one country and one language, and get each story's headline, source, date and link as JSON. **Quick answer — Is there an API for Google News results?** Google offers no public developer API for Google News search results. **SerpCrypto** fills that gap: set `engine=google_news` and the API returns the stories Google News shows for your query, country and language as JSON, each with headline, source, date, snippet and link. One successful search costs one credit, from $0.30/1K searches as of October 2026. No KYC; pay in crypto. ## Google News results with SerpCrypto | Attribute | Value | | --- | --- | | Endpoint | `GET https://serpcrypto.com/api/v1/search?engine=google_news&q=…` | | Array returned | `news_results`: `position`, `title`, `link`, `source`, `date`, `snippet`, `thumbnail` | | Freshness filter | `period=h` (past hour), `d` (day), `w` (week), `m` (month), `y` (year) | | Targeting | `gl` for the country edition, `hl` for the language | | Price | 1 credit per search; 25,000 searches for $18.75, 100,000 searches for $56.25 | | Payment | BTC, USDT, USDC, ETH, SOL, LTC; no KYC | ## What does a Google News search return? It returns a `news_results` array, one object per story, in the order Google News listed them. A story keeps its original link, so you read or cite the publisher's own page. | Field | Meaning | |---|---| | `position` | Rank in the list | | `title` | Headline as shown | | `link` | URL of the article on the publisher's site | | `source` | Publisher name | | `date` | Publication time as Google News displays it | | `snippet` | Short extract when shown | | `thumbnail` | URL of the small image when shown | The API returns headlines and links, not the full text of articles. Articles belong to their publishers; follow the link to read them. *Python: new stories about a brand in the past hour, without duplicates* ```python import json import os from pathlib import Path import requests API = "https://serpcrypto.com/api/v1/search" HEADERS = {"Authorization": "Bearer " + os.environ["SERPCRYPTO_API_KEY"]} SEEN = Path("seen-links.json") seen = set(json.loads(SEEN.read_text())) if SEEN.exists() else set() params = {"engine": "google_news", "q": '"Acme Robotics"', "gl": "us", "hl": "en", "period": "h"} resp = requests.get(API, headers=HEADERS, params=params, timeout=60) resp.raise_for_status() for story in resp.json().get("news_results", []): if story["link"] in seen: continue seen.add(story["link"]) print(story.get("date", ""), "|", story.get("source", ""), "|", story["title"]) print(" ", story["link"]) SEEN.write_text(json.dumps(sorted(seen))) ``` ## Who uses a Google News API? Teams that must know quickly when something is published about a name or a topic use it. The most common cases are these: - **Brand and PR monitoring:** every new story that mentions a company, a product or a spokesperson. See [brand monitoring](https://serpcrypto.com/use-cases/brand-monitoring/). - **News agents:** an AI agent that answers "what happened today about X" needs fresh, dated sources it can cite. See [AI agents](https://serpcrypto.com/use-cases/ai-agents/). - **Market and policy watch:** a sector, a regulation or a competitor, checked every morning in several countries. - **Research datasets:** how coverage of a topic changes over weeks, by publisher and by country edition. ## How do I get news from one country or in one language? Set `gl` to the country and `hl` to the language. `gl=de&hl=de` returns the German edition, `gl=in&hl=en` the English-language edition for India, and `gl=br&hl=pt-BR` Brazilian Portuguese. The same story can rank very differently between editions, so a global brand usually checks its three to five largest markets separately. Each edition is one search. The [locations page](https://serpcrypto.com/docs/locations/) lists the codes. ## How often should I check, and what does it cost? Check as often as you need to react. Hourly checks with `period=h` suit crisis watch and trading desks; a daily check with `period=d` suits most brand reports. The maths is simple: searches per month = queries × editions × checks per day × 30. Twenty brand and product names in one edition checked every hour is 20 × 24 × 30 = 14,400 searches a month. The same set checked once a day is 600. 25,000 searches ($18.75) cover the hourly setup for more than a month, and credits never expire. ## How much does the Google News engine cost? Google News costs one credit per successful search, like every other engine. 10,000 searches cost $7.50 and 100,000 searches cost $56.25, as of October 2026. | Plan | Quantity | Price | Price per unit | | --- | --- | --- | --- | | Test | 10,000 searches | $7.50 | $0.75/1K searches | | Starter | 25,000 searches | $18.75 | $0.75/1K searches | | Build | 100,000 searches | $56.25 | $0.56/1K searches | In every plan: Every engine at 1 credit per search · Failed requests not charged · Credits never expire · Country, language, city and device targeting **Before you pay** - Minimum top-up $25. Top up once in crypto, then buy any plan from your balance. - Unused balance stays on your account and never expires, and it pays for search credits. When a purchase costs less than the minimum top-up, the part added only to reach the minimum is refundable on request to support within 30 days of the payment, in USDT or USDC, as long as it has not been spent. The rest of the balance is not paid back, except where the law requires it. - Credited once your payment is checked on the blockchain. Typical network confirmation time: BTC ≈ 20 min · USDT TRC-20 ≈ 1 min · USDT ERC-20 ≈ 5 min · USDC ERC-20 ≈ 5 min · ETH ≈ 5 min · SOL ≈ 1 min · LTC ≈ 15 min ## Google News API: common questions ### Does Google have an official Google News API? No. Google does not offer a public developer API for Google News search results. Independent SERP APIs such as SerpCrypto run the search and return the results page as structured JSON. ### Can I get the full text of each article? No. The API returns the headline, source, date, snippet, thumbnail and link. The article itself belongs to its publisher; follow `link` to read it under the publisher's terms. ### How fresh are the results? Searches run at request time, so you get what Google News lists at that moment. Add `period=h` to keep only stories from the past hour, or `period=d` for the past day. ### Can an AI agent cite the stories it finds? Yes. Pass `title`, `source`, `date` and `link` to the model and ask it to cite the link it relies on. The [AI agents page](https://serpcrypto.com/use-cases/ai-agents/) shows how to keep the context small. ### How do I pay? Top up from $25 in BTC, USDT, USDC, ETH, SOL, LTC, then buy a pack from your balance. No ID, no selfie, no company documents and no card. --- SerpCrypto: SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. Prepaid search credits from $0.30/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto. --- # Google Search API: the full results page as JSON > Google Search API that returns the full results page as JSON: organic, ads, People Also Ask, AI Overview, local pack. From $0.30/1K searches, no KYC, crypto. - URL: https://serpcrypto.com/google-search-api/ - Last updated: 2026-10-22 - Publisher: SerpCrypto (https://serpcrypto.com) Send a query with a country, a language and, if you need it, a city. Get back what Google showed for it: organic results, ads, People Also Ask, the AI Overview, the local pack and more, as structured JSON. **Quick answer — Is there a Google Search API that returns the results page as JSON?** Yes. The **SerpCrypto** Google Search API runs a Google search for your query, country, language, city and device, and returns the results page as JSON: organic results, ads, People Also Ask, the AI Overview with its sources when Google shows one, the local pack, the knowledge panel and top stories. One successful search costs one credit, from $0.30/1K searches as of October 2026. No KYC; pay in crypto. ## Google Search API at a glance | Attribute | Value | | --- | --- | | Endpoint | `GET https://serpcrypto.com/api/v1/search?engine=google&q=…` | | Price | $7.50 for 10,000 searches, $56.25 for 100,000 searches, from $0.30/1K searches at 5,000,000 searches | | Billing | 1 credit = 1 successful search request, which returns one page of results on any supported engine. A request that fails is not charged | | Credit expiry | Never | | Targeting | Country (gl), language (hl), Google domain and a city-level location string, on desktop or mobile | | Results per page | About 10 organic results per page, as Google has served since it stopped honouring num=100 in September 2025; each extra page is one more credit | | Blocks returned | `organic_results`, `ads`, `people_also_ask`, `ai_overview`, `local_results`, `knowledge_graph`, `top_stories`, `inline_videos`, `shopping_results`, `related_searches`, `pagination` | | Authentication | An API key in the Authorization header (Bearer) or in the api_key query parameter | | Payment | BTC, USDT, USDC, ETH, SOL, LTC, from a $25 top-up; no KYC | ## What does the Google Search API return? It returns one JSON object per results page, with one key per block that Google showed. A block that is not on the page is simply absent, so your code should test for the key before reading it. | Key | What it holds | |---|---| | `organic_results` | Position, title, link, displayed link, snippet, date when shown, sitelinks | | `ads` | Text ads above and below the results, with their position and block | | `people_also_ask` | Each question, its answer snippet and the source link | | `ai_overview` | The AI Overview text and the pages it cites, when Google shows one | | `local_results` | The map pack: business name, address, rating, review count, category | | `knowledge_graph` | Title, type, description and source of the knowledge panel | | `top_stories` | News cards with title, link, source and date | | `related_searches` | The queries Google suggests at the bottom of the page | Every response starts with `search_metadata` (status, engine, time, `credits_used`) and `search_parameters` (the settings actually used). The [response format](https://serpcrypto.com/docs/response/) page lists every field with an example. *Python: top 10 links for a query in Germany* ```python import os import requests resp = requests.get( "https://serpcrypto.com/api/v1/search", headers={"Authorization": "Bearer " + os.environ["SERPCRYPTO_API_KEY"]}, params={"engine": "google", "q": "e-bike kaufen", "gl": "de", "hl": "de", "google_domain": "google.de"}, timeout=60, ) resp.raise_for_status() data = resp.json() for r in data.get("organic_results", []): print(r["position"], r["title"], r["link"]) print("credits used:", data["search_metadata"]["credits_used"]) ``` ## How do I get results for a specific country, city or device? Set `gl` to the two-letter country code, `hl` to the interface language, and `location` to a city-level place name such as `Austin, Texas, United States`. Add `device=mobile` to get the phone layout, which often orders local and video blocks differently from desktop. - **Country:** `gl=gb` returns what a searcher in the United Kingdom sees; pair it with `google_domain=google.co.uk`. - **Language:** `hl=es` with `gl=us` shows the Spanish-language interface in the United States. - **City:** `location` matters most for local intent queries such as plumbers, restaurants or clinics. - **Device:** `desktop` is the default; a mobile check of the same query is a separate search with its own credit. The [locations and languages](https://serpcrypto.com/docs/locations/) page has the codes for common markets and the format of location strings. ## How many results does one request return? One request returns one results page, which is about 10 organic results. Google stopped honouring the `num=100` parameter around 10 to 14 September 2025, so a top-100 check now takes 10 requests instead of one. A Search Engine Land analysis of 319 sites found that 87.7% of them lost Search Console impressions after the change, mostly from pages beyond the first results page. With SerpCrypto, `page=2` to `page=10` fetch the following pages, and each page is one credit. A daily top-10 check of 1,000 keywords is therefore 30,000 searches a month; a daily top-100 check is 300,000. The [rank tracking guide](https://serpcrypto.com/use-cases/rank-tracking/) shows how to budget both. ## Does the API return Google AI Overviews? Yes, when Google shows one. The `ai_overview` object holds the overview text and a `references` list with the title, link and source of each page it cites. Google does not show an AI Overview for every query, country or device, so the key is absent on many results pages. AI Overview data answers two practical questions: whether your pages are cited for a query, and which competitors are. Store `ai_overview.references` next to `organic_results` positions, and you can follow both over time with the same daily job. ## Do Google SERP APIs still work after the 2025 and 2026 changes? They work, but every provider, including us, deals with more friction than two years ago. These are the public facts as of 22 October 2026: - **January 2025:** Google started requiring JavaScript for Search, which ended plain HTTP fetching of results pages. - **September 2025:** `num=100` stopped working, so deep checks cost 10 times more requests. - **July–August 2026:** result links started going through a server-side `/goto` redirect. - **Since about 11 September 2026:** rank-tracking vendors reported wide blocks on large residential proxy networks, more CAPTCHAs and altered results. Nozzle measured 8.0% of results pages affected; DemandSphere about 11%. What this means for you: a search that fails returns an error and **is not charged**, so failures cost you time, not credits. We do not publish a success rate or a response time until we have measured them over a public sample; when we do, the method and date will be on this page. ## How much does the Google Search API cost? Google costs the same single credit as every other engine. 10,000 searches cost $7.50, 100,000 searches cost $56.25 and 1,000,000 searches cost $375, as of October 2026. Credits never expire. See [all seven packs](https://serpcrypto.com/pricing/). | Plan | Quantity | Price | Price per unit | | --- | --- | --- | --- | | Test | 10,000 searches | $7.50 | $0.75/1K searches | | Build | 100,000 searches | $56.25 | $0.56/1K searches | | Scale | 500,000 searches | $281.25 | $0.56/1K searches | | Volume | 1,000,000 searches | $375 | $0.38/1K searches | In every plan: Every engine at 1 credit per search · Failed requests not charged · Credits never expire · Country, language, city and device targeting **Before you pay** - Minimum top-up $25. Top up once in crypto, then buy any plan from your balance. - Unused balance stays on your account and never expires, and it pays for search credits. When a purchase costs less than the minimum top-up, the part added only to reach the minimum is refundable on request to support within 30 days of the payment, in USDT or USDC, as long as it has not been spent. The rest of the balance is not paid back, except where the law requires it. - Credited once your payment is checked on the blockchain. Typical network confirmation time: BTC ≈ 20 min · USDT TRC-20 ≈ 1 min · USDT ERC-20 ≈ 5 min · USDC ERC-20 ≈ 5 min · ETH ≈ 5 min · SOL ≈ 1 min · LTC ≈ 15 min ## Google Search API: common questions ### Is this Google's official API? No. SerpCrypto is an independent SERP API: it runs the search and parses the page Google returns. Google's own developer search product is the Custom Search JSON API, which searches a configured set of sites; see how the two compare on our [Google Custom Search API alternative](https://serpcrypto.com/alternatives/google-custom-search-api/) page. ### Can I get more than 10 results in one request? No. Google stopped honouring `num=100` in September 2025, so one results page holds about 10 organic results. Request `page=2` to `page=10` for positions 11 to 100; each page is one credit. ### Are failed searches charged? No. 1 credit = 1 successful search request, which returns one page of results on any supported engine. A request that fails is not charged. Errors come back with an HTTP status and a code such as `search_failed` or `search_timeout`; the [errors page](https://serpcrypto.com/docs/errors/) shows a retry loop with backoff. ### Can I search Google Maps, News, Images or Shopping too? Yes, with the same key and the same credits: set `engine` to `google_maps`, `google_news`, `google_images` or `google_shopping`. Each has its own page: [Maps](https://serpcrypto.com/google-maps-api/), [News](https://serpcrypto.com/google-news-api/), [Images](https://serpcrypto.com/google-images-api/), [Shopping](https://serpcrypto.com/google-shopping-api/). ### Do I need a company or an ID to get a key? No. SerpCrypto is a no-KYC SERP API: no ID, no selfie, no company documents and no card. Account-number login; email is optional. You top up from $25 in BTC, USDT, USDC, ETH, SOL, LTC. ## Sources 1. [Google begins requiring JavaScript for Google Search](https://techcrunch.com/2025/01/17/google-begins-requiring-javascript-for-google-search/) — TechCrunch (2025-01-17) 2. [Google num=100 impact data](https://searchengineland.com/google-num100-impact-data-462231) — Search Engine Land (2025-09) 3. [Google's goto redirect](https://susodigital.com/thoughts/googles-goto-redirect/) — SUSO Digital (2026-09-25) 4. [Third-party Google rank tracking broken](https://www.seroundtable.com/third-party-google-rank-tracking-broken-42272.html) — Search Engine Roundtable (2026-10-12) 5. [Google returns wrong results on 11% of DemandSphere's daily rank checks](https://ppc.land/google-returns-wrong-results-on-11-of-demandspheres-daily-rank-checks) — PPC Land (2026-10-13) --- SerpCrypto: SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. Prepaid search credits from $0.30/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto. --- # Google Shopping API: product offers and prices as JSON > Google Shopping API for price monitoring: offers with price, merchant, delivery and rating as JSON, by country, from $0.30/1K searches. No KYC, crypto. - URL: https://serpcrypto.com/google-shopping-api/ - Last updated: 2026-10-22 - Publisher: SerpCrypto (https://serpcrypto.com) Search Google Shopping for a product in a given country and read every offer it lists: title, price, merchant, delivery line, rating and link, ready for a price sheet or a database. **Quick answer — Is there an API that returns Google Shopping results and prices?** Yes. With **SerpCrypto**, set `engine=google_shopping` and the API returns the offers Google Shopping lists for your query and country as JSON: title, price as text and as a number, currency, merchant, delivery line, rating, review count and link. One successful search costs one credit, from $0.30/1K searches as of October 2026. No KYC; pay in crypto. ## Google Shopping results with SerpCrypto | Attribute | Value | | --- | --- | | Endpoint | `GET https://serpcrypto.com/api/v1/search?engine=google_shopping&q=…` | | Array returned | `shopping_results` | | Fields | `position`, `title`, `price`, `extracted_price`, `currency`, `source`, `link`, `delivery`, `rating`, `reviews` | | Targeting | `gl` for the country (and so the currency), `hl` for the language, `location` for a city | | Price | 1 credit per search; 100,000 searches for $56.25 | | Payment | BTC, USDT, USDC, ETH, SOL, LTC; no KYC, no card | ## What does a Google Shopping search return? It returns a `shopping_results` array, one object per offer, in the order Google Shopping showed them. Two price fields save you parsing work: `price` keeps the text as displayed, and `extracted_price` gives the number. | Field | Meaning | |---|---| | `position` | Rank of the offer in the list | | `title` | Product title as the merchant wrote it | | `price` | Price text as shown, for example "$129.99" | | `extracted_price` | The same price as a number | | `currency` | Currency of the price | | `source` | Merchant name | | `link` | Link to the offer | | `delivery` | Delivery line when shown, such as free shipping | | `rating`, `reviews` | Product rating and number of reviews, when shown | *Python: price range and merchants for one product in two countries* ```python import os import requests API = "https://serpcrypto.com/api/v1/search" HEADERS = {"Authorization": "Bearer " + os.environ["SERPCRYPTO_API_KEY"]} def offers(query, gl, hl): params = {"engine": "google_shopping", "q": query, "gl": gl, "hl": hl} resp = requests.get(API, headers=HEADERS, params=params, timeout=60) resp.raise_for_status() return [o for o in resp.json().get("shopping_results", []) if o.get("extracted_price")] for gl, hl in [("us", "en"), ("de", "de")]: found = offers("sony wh-1000xm5", gl, hl) if not found: print(gl, "no priced offers") continue prices = [o["extracted_price"] for o in found] merchants = sorted({o.get("source", "?") for o in found}) print(gl, found[0].get("currency", ""), min(prices), "to", max(prices), "-", ", ".join(merchants[:5])) ``` ## How is this different from Google's Merchant API? Google's Merchant API is for merchants who manage their own products. Google describes it as automating "account, product, and inventory management within Merchant Center". It works on your catalogue, not on what shoppers see. The SerpCrypto Shopping engine reads the public results page instead. It shows every merchant that Google lists for a query, with their prices, so you can compare your offer with theirs. Many teams use both: the Merchant API to publish their products, and a Shopping results API to watch the market. ## How do I compare prices between countries? Run the same product query with a different `gl` for each country, and the matching `hl`. The country decides which merchants and currency appear, so `gl=gb` returns offers in pounds from UK shops, and `gl=fr&hl=fr` offers in euros from French ones. Compare `extracted_price` within one country first. Across countries, convert with your own exchange rates and keep the `delivery` line in mind, because shipping and taxes change the real price a shopper pays. ## How many searches does price monitoring need? Count one search per product query, country and check. Watching 500 products in 3 countries once a day is 500 × 3 × 30 = 45,000 searches a month. Twice a day doubles it to 90,000. 100,000 searches ($56.25) cover the daily setup for about two months. Credits never expire, so you can slow down after a sales season without losing what you bought. The [price monitoring guide](https://serpcrypto.com/use-cases/price-monitoring/) covers matching, alerts and storage. ## How much does the Google Shopping engine cost? Shopping searches cost one credit each, like every other engine. 100,000 searches cost $56.25 and 500,000 searches cost $281.25, as of October 2026. | Plan | Quantity | Price | Price per unit | | --- | --- | --- | --- | | Starter | 25,000 searches | $18.75 | $0.75/1K searches | | Build | 100,000 searches | $56.25 | $0.56/1K searches | | Grow | 250,000 searches | $140.63 | $0.56/1K searches | | Scale | 500,000 searches | $281.25 | $0.56/1K searches | In every plan: Every engine at 1 credit per search · Failed requests not charged · Credits never expire · Country, language, city and device targeting **Before you pay** - Minimum top-up $25. Top up once in crypto, then buy any plan from your balance. - Unused balance stays on your account and never expires, and it pays for search credits. When a purchase costs less than the minimum top-up, the part added only to reach the minimum is refundable on request to support within 30 days of the payment, in USDT or USDC, as long as it has not been spent. The rest of the balance is not paid back, except where the law requires it. - Credited once your payment is checked on the blockchain. Typical network confirmation time: BTC ≈ 20 min · USDT TRC-20 ≈ 1 min · USDT ERC-20 ≈ 5 min · USDC ERC-20 ≈ 5 min · ETH ≈ 5 min · SOL ≈ 1 min · LTC ≈ 15 min ## Google Shopping API: common questions ### Can I search by GTIN, EAN or product ID? Search with the text a shopper would type, such as brand and model. A barcode number sometimes works as a query, but Google decides what it matches. Store the offer titles you get back and match them to your catalogue. ### Are prices shown with tax and shipping? The API returns the price exactly as Google Shopping displays it in that country, plus the `delivery` line when there is one. Whether tax is included depends on the country and the merchant. ### Can I monitor one merchant only? Filter the results on `source`, the merchant name. To see a merchant's position for many products, run each product query and keep the offers from that merchant. ### Is this Google's official Shopping API? No. SerpCrypto is an independent SERP API. Google's Merchant API is a separate product for merchants who manage their own listings in Merchant Center. ### How do I pay? Top up from $25 in BTC, USDT, USDC, ETH, SOL, LTC and buy a pack from your balance. SerpCrypto asks for no ID, no selfie and no card. ## Sources 1. [Overview of Merchant API](https://developers.google.com/merchant/api/overview) — Google for Developers (checked 2026-10-22) --- SerpCrypto: SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. Prepaid search credits from $0.30/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto. --- # Does Google have a search API in 2026? > Does Google have a search API in 2026? Custom Search JSON API closed to new customers, full-web search ends 1 January 2027. Every option, price and date. - URL: https://serpcrypto.com/guides/does-google-have-a-search-api/ - Last updated: 2026-10-22 - Publisher: SerpCrypto (https://serpcrypto.com) Google's developer search products have narrowed fast since January 2026. Here is what each one still offers, who can use it, what it costs and when it ends. **Quick answer — Does Google have an official search API in 2026?** Not one that a new developer can sign up for to search the whole web. As of 22 October 2026, Google's Custom Search JSON API is closed to new customers and ends on 1 January 2027, and new Programmable Search Engines are limited to 50 domains. Full-web access is partner-only. Third-party SERP APIs such as **SerpCrypto** fill the gap, from $0.30/1K searches, paid in crypto, no KYC. ## Which Google search options exist, and who can use them? Read on Google's own pages and in Search Engine Journal on 22 October 2026. A dash means the source does not give the figure. | Option | Who can use it | Searches the whole web? | Price | End date | | --- | --- | --- | --- | --- | | Custom Search JSON API | Existing customers only | Only through an engine still set to the entire web | 100 queries a day free, then $5 per 1,000, up to 10,000 a day | 1 January 2027 | | Programmable Search Engine | Anyone, for new engines with sites to search | New engines: no, up to 50 domains. Existing full-web engines: until 1 January 2027 | — | Full-web option ends 1 January 2027 | | Vertex AI Search | Google Cloud customers | Google presents it for searching up to 50 domains | Google Cloud pricing | — | | Web Search Service API | Partners with a partner client ID | Yes | No public price | — | | Third-party SERP API (e.g. SerpCrypto) | Anyone | Yes: the page Google shows a searcher | Per 1,000 searches; SerpCrypto from $0.30/1K searches | — | *Prices checked on October 22, 2026.* ## What happened to the Custom Search JSON API? Google closed it to new customers and set an end date. Its overview page, read on 22 October 2026, states that "The Custom Search JSON API is closed to new customers" and that existing customers "have until January 1, 2027 to transition to an alternative solution". Until then, existing customers keep the same terms: 100 queries a day at no cost, then $5 per 1,000 queries, with a ceiling of 10,000 queries a day. The API queries a Programmable Search Engine, so what it can search depends on how that engine is configured (see below). ## Can Programmable Search Engine still search the entire web? Only engines created before the change, and only until 1 January 2027. Google's help page says that as of 20 January 2026, Programmable Search Engine "is transitioning to more focused solutions": new engines must list the sites to search, up to 50 domains. Existing engines set to "Search the entire web" can keep that setting until 1 January 2027. The same page warns that once the option is switched off, it cannot be switched back on. Google announced the change on its Programmable Search Engine blog on 20 January 2026. ## What does Google recommend instead? For site search, Google points to Vertex AI Search, which its Custom Search page describes as "a favorable alternative for searching up to 50 domains". For needs that cover the whole web, the same page asks developers to contact Google rather than naming a self-service product. In practice, this suits a company that wants search over its own sites and a few partners. It does not give a rank tracker, an SEO tool or an AI agent the results page that a person sees on google.com. ## What is Google's Web Search Service API? It is a full-web results API that Google documents for partners only. Search Engine Journal reported on 11 September 2026 that Google had updated its documentation on 9 September 2026. According to that report, access needs a Google Cloud project, an API key and a partner client ID, and each request must include the end user's IP address. A request returns up to 20 results, 10 by default. No public price was given, and the overview page at developers.google.com returned an error when we tried to read it on 22 October 2026. ## What can developers use for full Google results now? A third-party SERP API. It runs the search for the country, language, city and device you choose, and returns the results page as JSON, including blocks no official API ever returned, such as People Also Ask, the local pack and the AI Overview. **SerpCrypto** is one of them: `GET https://serpcrypto.com/api/v1/search?engine=google&q=…`, 1 credit = 1 successful search request, which returns one page of results on any supported engine. A request that fails is not charged. 10,000 searches cost $7.50 and 100,000 searches cost $56.25, as of October 2026, and credits never expire. If you are leaving the Custom Search JSON API, our [migration page](https://serpcrypto.com/alternatives/google-custom-search-api/) maps its parameters to ours. Automated access to Google results is contested in US courts in 2026; our page [Is using a SERP API legal?](https://serpcrypto.com/guides/is-a-serp-api-legal/) gives the dated facts. ## And what about Bing? Bing no longer has a results API either. Microsoft retired the Bing Search APIs on 11 August 2025 and decommissioned existing instances. Its suggested replacement, Grounding with Bing Search for Azure AI Agents, gives a model grounded answers rather than a results page, and Microsoft lists it at $14 per 1,000 transactions. Details are on our [Bing Search API replacement](https://serpcrypto.com/alternatives/bing-search-api/) page. ## Google search API: common questions ### Is the Google Custom Search API deprecated? Yes, in effect. Google closed it to new customers and gave existing customers until 1 January 2027 to move to another solution. Until that date it still works for them at $5 per 1,000 queries after 100 free queries a day. ### Is there a free Google search API? Only for existing Custom Search JSON API customers, who get 100 queries a day at no cost until 1 January 2027. New developers cannot sign up. Third-party SERP APIs charge per search. ### What are the Custom Search JSON API limits? Existing customers can run up to 10,000 queries a day: the first 100 are free and the rest cost $5 per 1,000. What the queries cover depends on the Programmable Search Engine behind them. ### Can I still create a Programmable Search Engine that searches the entire web? No. Since 20 January 2026, new engines must list the sites to search, up to 50 domains. Only engines that already had the full-web option can keep it, until 1 January 2027. ### Does SerpCrypto need a Google account or an API key from Google? No. You use a SerpCrypto API key only. You top up from $25 in BTC, USDT, USDC, ETH, SOL, LTC, buy a credit pack, and call the endpoint; no Google Cloud project is involved. ## Sources 1. [Custom Search JSON API overview](https://developers.google.com/custom-search/v1/overview) — Google for Developers (read 2026-10-22) 2. [Programmable Search Engine: search the entire web](https://support.google.com/programmable-search/answer/12397162?hl=en) — Google Help (read 2026-10-22) 3. [Updates to our Web Search Products & Programmable Search Engine Capabilities](https://programmablesearchengine.googleblog.com/2026/01/updates-to-our-web-search-products.html) — Programmable Search Engine Blog (2026-01-20) 4. [Google documents partner-only API for full web search results](https://www.searchenginejournal.com/google-documents-partner-only-api-for-full-web-search-results/) — Search Engine Journal (2026-09-11) 5. [Bing Search API retirement](https://learn.microsoft.com/en-us/lifecycle/announcements/bing-search-api-retirement) — Microsoft Learn (read 2026-10-22) 6. [Grounding with Bing Search pricing](https://www.microsoft.com/en-us/bing/apis/grounding-pricing) — Microsoft (read 2026-10-22) --- SerpCrypto: SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. Prepaid search credits from $0.30/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto. --- # Google removed 100 results per page: what changed, and what it costs > Google stopped serving 100 results per page in September 2025. The num=100 change, /goto links and 2026 blocking, with dates and what they cost API buyers. - URL: https://serpcrypto.com/guides/google-100-results-change/ - Last updated: 2026-10-22 - Publisher: SerpCrypto (https://serpcrypto.com) Three changes in thirteen months made Google results harder and more expensive to collect: the end of num=100, redirect links, and a wave of blocking in September 2026. Here is the dated timeline and what it means for a SERP API budget. **Quick answer — Did Google remove 100 results per page?** Yes. Around 10 to 14 September 2025, Google stopped honouring the `num=100` parameter, so a results page now holds about 10 organic results and a top-100 check takes 10 requests instead of one. With **SerpCrypto**, each page costs one credit, from $0.30/1K searches as of October 2026, and failed requests are not charged. No KYC; pay in crypto. ## The timeline in five lines | Attribute | Value | | --- | --- | | 17 January 2025 | Google Search starts requiring JavaScript | | About 14 September 2025 | `num=100` stops working; 100 results need 10 requests | | 26 August 2026 | Google confirms server-side `/goto` redirect links in results | | About 11–13 September 2026 | Rank-tracking vendors report wide blocking aimed at large residential proxy networks | | Since 13 September 2026 | Nozzle measures 8.0% of results pages with swapped results, against 0.17% before | ## What exactly happened to num=100? Google stopped returning 100 results when a URL asked for them with `num=100`. PPC Land dates the change to 14 September 2025; other reports place the first signs from 10 September. Search Engine Land reported that Google described the parameter as not officially supported. For anyone tracking positions beyond the first page, the effect was mechanical: where one request used to return the top 100, ten requests are now needed. A Search Engine Land analysis of 319 sites found that 87.7% of them lost Search Console impressions after the change, which suggests that many recorded impressions had come from automated deep checks. ## What are Google's /goto links? They are redirect links: result URLs now pass through `google.com/goto` before reaching the site. PPC Land reports a first public sighting on 23 June 2026, and Nozzle had seen them on and off for about four months before that. Google confirmed the change on 26 August 2026, saying it has "a long history of deploying technical measures against evolving forms of abuse". Nozzle reported a rollout close to 100% across several residential IP providers. Derek Perkins of Nozzle estimated that resolving a five-page ranking now takes 500 to 1,000 requests for a tool that collects pages itself. ## What happened in September and October 2026? Collection got harder again, and some returned pages became unreliable. DemandSphere said that "right around September 11th" Google deployed "the most aggressive blocking that we have seen to date". Nozzle saw request failures from 13 September. Derek Perkins described three parts: wide IP blocks aimed at major residential proxy networks, more challenge pages, and wrong results served to traffic that looks automated. Earlier, on 2 and 3 July 2026, Google's threat intelligence team and partners including the FBI had disrupted the NetNut residential proxy network, according to BleepingComputer. Google has made no statement about the swapped results. ## How many results pages were affected? Figures published by rank-tracking vendors about their own data. Each measures differently, so they are not directly comparable. | Source | What it measured | Figure | Period | | --- | --- | --- | --- | | Nozzle | Results pages with swapped results | 8.0%, against 0.17% before; peaks of 24.9–27.0% | Since 13 September 2026; peaks 27 and 29 September | | Nozzle | Same, daily | 5–11% per day | Since 1 October 2026, 27.5 million pages in 49 countries | | AccuRanker | Affected checks | About 1.47% | 1 to 9 October 2026 | | DemandSphere | Daily rank checks with wrong results | About 11% | 7 October 2026 | *Prices checked on October 22, 2026.* ## What does this mean for a SERP API budget? Depth now drives cost. With about 10 results per page, tracking the top 10 costs one search per keyword and check, the top 30 three searches, and the top 100 ten. A daily top-100 check of 1,000 keywords is 300,000 searches a month. On **SerpCrypto**, 1 credit = 1 successful search request, which returns one page of results on any supported engine. A request that fails is not charged. 100,000 searches cost $56.25 and 1,000,000 searches cost $375, as of October 2026, and credits never expire. Most teams cut the bill by stopping once they find their domain, and by checking long-tail keywords weekly; the [rank tracking page](https://serpcrypto.com/use-cases/rank-tracking/) has the code and a cost table. ## How should I treat results that look wrong? Treat a sudden jump as a question, not a fact. Since September 2026 several vendors have published evidence that some pages served to automated traffic differ from what people see. - **Re-check outliers:** run the same query again later, or from another nearby location, before reporting a big move. - **Compare devices:** PPC Land reported on 19 October 2026 that detection tightened again on 15 October, mostly on mobile. - **Keep the raw JSON:** with the full page stored, you can audit a suspicious day later without paying for it again. SerpCrypto does not publish a success rate or a data-quality figure until we have measured them on a public sample. When we do, the method and date will be on this page. ## The 100-results change: common questions ### Can I still get 100 Google results in one request? No. Since September 2025 Google serves about 10 organic results per page whatever the request asks for. Fetch `page=1` to `page=10` to cover the top 100; on SerpCrypto each page is one credit. ### Why did my Search Console impressions drop in September 2025? Many impressions had come from tools that loaded 100 results at once. When `num=100` stopped working, those deep, automated impressions disappeared. A Search Engine Land analysis of 319 sites found that 87.7% lost impressions. ### Are failed searches charged on SerpCrypto? No. A request that fails returns an error code such as `search_failed` or `search_timeout`, and it costs no credit. The [errors page](https://serpcrypto.com/docs/errors/) has a retry loop. ### Did Google explain the September 2026 blocking? On the `/goto` links, Google said it has a long history of technical measures against abuse. On the swapped results reported since September 2026, it had made no statement as of 22 October 2026. ### How do I budget a top-100 tracker now? Multiply keywords by 10 pages and by checks per month. 1,000 keywords checked weekly to position 100 is about 40,000 searches a month, which 100,000 searches ($56.25) covers for about two and a half months. ## Sources 1. [Google begins requiring JavaScript for Google Search](https://techcrunch.com/2025/01/17/google-begins-requiring-javascript-for-google-search/) — TechCrunch (2025-01-17) 2. [Google forces rank trackers into 500 to 1,000 requests per query](https://ppc.land/google-forces-rank-trackers-into-500-to-1-000-requests-per-query/) — PPC Land (2026-08-27) 3. [Google Search confirms it does not support the results per page parameter (reported)](https://searchengineland.com/google-search-confirms-it-does-not-support-the-results-per-page-parameter-462244) — Search Engine Land (2025-09) 4. [Google num=100 impact data](https://searchengineland.com/google-num100-impact-data-462231) — Search Engine Land (2025-09) 5. [NetNut proxy network disrupted, 2 million infected devices cut off](https://bleepingcomputer.com/news/security/netnut-proxy-network-disrupted-2-million-infected-devices-cut-off) — BleepingComputer (2026-07-03) 6. [Google returns wrong results on 11% of DemandSphere's daily rank checks](https://ppc.land/google-returns-wrong-results-on-11-of-demandspheres-daily-rank-checks/) — PPC Land (2026-10-13) 7. [Third-party Google rank tracking broken](https://www.seroundtable.com/third-party-google-rank-tracking-broken-42272.html) — Search Engine Roundtable (2026-10-12) 8. [Google feeds rank trackers swapped results on 8% of pages, Nozzle finds](https://ppc.land/google-feeds-rank-trackers-swapped-results-on-8-of-pages-nozzle-finds/) — PPC Land (2026-10-19) --- SerpCrypto: SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. Prepaid search credits from $0.30/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto. --- # Google AI Overview API and AI Mode API in 2026 > Is there a Google AI Overview API in 2026? What Google offers, how SERP APIs return AI Overview text and sources, and code to track citations of your pages. - URL: https://serpcrypto.com/guides/google-ai-overview-api/ - Last updated: 2026-10-22 - Publisher: SerpCrypto (https://serpcrypto.com) AI Overviews now sit above the results for a large share of searches. Here is how to get their text and sources as data, what is and is not available for AI Mode, and how to track whether your pages are cited. **Quick answer — Is there a Google AI Overview API?** Google publishes no API that returns AI Overviews as they appear in Search, as far as we found on 22 October 2026. SERP APIs fill the gap: **SerpCrypto** returns the AI Overview text and its cited sources in the `ai_overview` object whenever Google shows one, for one credit per search from $0.30/1K searches. SerpCrypto has no AI Mode engine. No KYC; pay in crypto. ## AI Overviews and AI Mode: the key facts | Attribute | Value | | --- | --- | | Share of searches with an AI Overview | 43%, up from 15% a year earlier (Similarweb data, TechCrunch, 27 July 2026; geography not stated) | | AI Mode visits | 279 million in May 2026, against 126 million in June 2025 (same source) | | Official Google API | None found for AI Overviews or AI Mode | | SerpCrypto field | `ai_overview`: `text` and `references` (index, title, link, source) | | SerpCrypto AI Mode | Not offered | | Cost | 1 credit = 1 successful search request, which returns one page of results on any supported engine. A request that fails is not charged | ## Why does AI Overview data matter in 2026? Because a growing share of searches show one. Similarweb data reported by TechCrunch on 27 July 2026 put AI Overviews on 43% of searches, up from 15% a year earlier. The same data counted 279 million AI Mode visits in May 2026, against 126 million in June 2025. When an overview appears, it sits above the organic results and cites a handful of pages. A site can rank on page one and still be missing from that list, or be cited while ranking lower. Positions alone no longer describe visibility. ## What does the ai_overview object contain? It holds the overview text and the list of pages the overview cites. On **SerpCrypto**, the object sits next to `organic_results` in the same response, so one search gives you both views of the page. | Field | Content | |---|---| | `text` | The overview as plain text | | `references[].index` | Order of the reference in the overview | | `references[].title` | Title of the cited page | | `references[].link` | URL of the cited page | | `references[].source` | Site name shown by Google | The key is absent when Google shows no overview for that query, place, language and device. Your code should treat a missing key as "no overview", not as an error. ## Do all SERP APIs capture AI Overviews equally well? No, and published tests show wide gaps. In a vendor benchmark of 25 identical queries, updated on 30 May 2026, Scrape.do reported AI Overview detection rates from 0% to 68% across the providers it tested, with SerpApi at 68%. Overviews can load after the rest of the page, and some depend on location and device. That test was run by a provider about itself and its rivals, so read it as a direction, not a ranking. SerpCrypto has not yet published its own measured rate; when it does, the method, sample and date will be on this page. ## Is there an AI Mode API? Not from Google, and not from SerpCrypto. AI Mode is a separate conversational search experience; we found no public Google API for it on 22 October 2026. Some SERP API providers sell AI Mode endpoints; SerpCrypto does not, and our engines are Google Search, Google News, Google Images, Google Maps (local results), Google Shopping and Bing. If AI Mode matters to your reporting, track it with a provider that offers it, and use SerpCrypto for the classic results page and its AI Overview. ## How do I track whether my pages are cited in AI Overviews? Run your keyword list through the API, read `ai_overview.references`, and record whether your domain is among them, next to your organic position. Do it on a fixed schedule, from the same location and device, so that changes reflect Google and not your settings. The script below writes one CSV row per keyword. *Python: AI Overview citations and organic position per keyword* ```python import csv import os from urllib.parse import urlparse import requests API = "https://serpcrypto.com/api/v1/search" HEADERS = {"Authorization": "Bearer " + os.environ["SERPCRYPTO_API_KEY"]} def host(url): return urlparse(url or "").netloc.lower().removeprefix("www.") def check(keyword, domain, gl="us", hl="en"): params = {"engine": "google", "q": keyword, "gl": gl, "hl": hl} resp = requests.get(API, headers=HEADERS, params=params, timeout=60) resp.raise_for_status() data = resp.json() overview = data.get("ai_overview") cited = [] if overview: cited = [host(r.get("link")) for r in overview.get("references", [])] position = next((r["position"] for r in data.get("organic_results", []) if host(r.get("link")) == domain), None) return { "keyword": keyword, "has_overview": bool(overview), "we_are_cited": domain in cited, "cited_domains": " ".join(cited), "organic_position": position, } keywords = ["how to descale a kettle", "best time to water tomatoes"] with open("ai_overview_tracking.csv", "w", newline="", encoding="utf-8") as fh: writer = csv.DictWriter(fh, fieldnames=["keyword", "has_overview", "we_are_cited", "cited_domains", "organic_position"]) writer.writeheader() for kw in keywords: writer.writerow(check(kw, "example.com")) ``` ## How much does AI Overview tracking cost? The overview comes with the normal results page, so it costs no extra credit. One keyword checked once is one search. A weekly check of 500 keywords is about 2,000 searches a month; 10,000 searches ($7.50) covers five months of it. A daily check of 2,000 keywords is 60,000 searches a month, which fits in 100,000 searches ($56.25). Prices as of October 2026; credits never expire. You pay in BTC, USDT, USDC, ETH, SOL, LTC, with no ID upload. ## AI Overview API: common questions ### Does every Google search have an AI Overview? No. Similarweb data reported in July 2026 put them on 43% of searches. Whether one appears depends on the query, the country, the language and the device, so the `ai_overview` key is often absent. ### Does the AI Overview cost extra on SerpCrypto? No. It is part of the Google results page, so it comes in the same response as the organic results, for the same single credit. ### Can I get AI Mode answers from SerpCrypto? No. SerpCrypto does not offer an AI Mode engine. It returns the classic Google results page, including the AI Overview when Google shows one, plus Google News, Images, Maps, Shopping and Bing. ### Can I see AI Overviews for another country or city? Yes. Set `gl`, `hl` and `location` as for any search. Overviews differ by market, so compare like with like: same country, language and device each time. ### Is the AI Overview text stored by SerpCrypto? We return it to you and do not sell it. We log, for each request, the time, account, engine, query parameters, status and credits used, and keep these logs for 30 days to show your usage and settle billing disputes, then delete them. We do not sell or share your queries. We keep your account number, orders, top-up records and credit totals while your account is open, and afterwards only as long as accounting rules require. We disclose records to authorities only when the law requires it. ## Sources 1. [Google's AI search is rapidly becoming the default, new data shows](https://techcrunch.com/2026/07/27/googles-ai-search-is-rapidly-becoming-the-default-new-data-shows/) — TechCrunch (2026-07-27) 2. [SerpApi alternatives (benchmark of 25 queries)](https://scrape.do/blog/serpapi-alternatives/) — Scrape.do (2026-05-30) --- SerpCrypto: SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. Prepaid search credits from $0.30/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto. --- # How to get Google search results in Python (2026) > How to get Google search results in Python in 2026: why plain requests fail, then tested code for requests, pandas, async httpx, pagination and JSONL files. - URL: https://serpcrypto.com/guides/google-search-results-python/ - Last updated: 2026-10-22 - Publisher: SerpCrypto (https://serpcrypto.com) Fetching google.com with requests stopped being an option in 2025. This guide explains why, then gives short, syntax-checked Python that gets the same results as JSON through a SERP API. **Quick answer — How do I get Google search results in Python in 2026?** Call a SERP API from Python and read the JSON it returns; do not fetch google.com yourself, because Google has required JavaScript since 17 January 2025 and its robots.txt disallows `/search`. With **SerpCrypto**, one `requests.get` to `https://serpcrypto.com/api/v1/search` returns organic results, People Also Ask and more, for one credit per search from $0.30/1K searches as of October 2026. No KYC; pay in crypto. ## What you need | Attribute | Value | | --- | --- | | Python | 3.9 or newer | | Packages | `requests` for the basics; `pandas` for tables; `httpx` for async | | API key | From your SerpCrypto dashboard, stored in the `SERPCRYPTO_API_KEY` environment variable | | Endpoint | `GET https://serpcrypto.com/api/v1/search` | | Cost | 1 credit = 1 successful search request, which returns one page of results on any supported engine. A request that fails is not charged | | Smallest pack | 10,000 searches for $7.50; credits never expire | ## Why doesn't requests.get on google.com work any more? Because the results are no longer in the HTML that a plain HTTP client receives. On 17 January 2025, Google started requiring JavaScript for Search; it said JavaScript helps it protect its services from bots and that fewer than 0.1% of searches came from users with JavaScript turned off, as TechCrunch reported. Google's robots.txt, read on 22 October 2026, also lists `Disallow: /search` for all user agents. A SERP API takes care of running the search and parsing the page, and your Python code only deals with JSON. That is the approach used in the rest of this guide. ## How do I make the first request? Install requests, put your key in an environment variable, and send the query with a country and a language. The response is a dictionary with one key per block of the results page; `organic_results` holds the ranked links. Keep the key out of your source code. On macOS and Linux, `export SERPCRYPTO_API_KEY=...` in the shell; on Windows PowerShell, `$env:SERPCRYPTO_API_KEY="..."`. *First request with requests* ```python import os import requests API = "https://serpcrypto.com/api/v1/search" KEY = os.environ["SERPCRYPTO_API_KEY"] resp = requests.get( API, headers={"Authorization": "Bearer " + KEY}, params={"engine": "google", "q": "best hiking boots", "gl": "us", "hl": "en"}, timeout=60, ) resp.raise_for_status() results = resp.json() for item in results.get("organic_results", []): print(item["position"], item["title"]) print(" ", item["link"]) for q in results.get("people_also_ask", []): print("PAA:", q["question"]) ``` ## How do I turn the results into a pandas DataFrame? Build one row per organic result and keep the query next to it. A DataFrame makes it easy to sort, filter by domain, or export to CSV and Excel. Missing fields, such as `date` on most results, become empty cells instead of errors. *Several queries into one DataFrame* ```python import os from urllib.parse import urlparse import pandas as pd import requests API = "https://serpcrypto.com/api/v1/search" HEADERS = {"Authorization": "Bearer " + os.environ["SERPCRYPTO_API_KEY"]} def organic_rows(query, gl="us", hl="en"): params = {"engine": "google", "q": query, "gl": gl, "hl": hl} resp = requests.get(API, headers=HEADERS, params=params, timeout=60) resp.raise_for_status() for r in resp.json().get("organic_results", []): yield { "query": query, "position": r.get("position"), "title": r.get("title"), "domain": urlparse(r.get("link", "")).netloc, "link": r.get("link"), "snippet": r.get("snippet"), } queries = ["crm for small business", "invoice software", "time tracking app"] df = pd.DataFrame([row for q in queries for row in organic_rows(q)]) print(df.groupby("domain").size().sort_values(ascending=False).head(10)) df.to_csv("serp_results.csv", index=False) ``` ## How do I run many searches at once? Use httpx with asyncio and cap the number of requests in flight with a semaphore. Your account has a concurrency limit, shown in the dashboard; staying under it avoids `429` errors. Start with a small number, such as 5, and set it to the limit your dashboard shows. *Async searches with httpx and a semaphore* ```python import asyncio import os import httpx API = "https://serpcrypto.com/api/v1/search" HEADERS = {"Authorization": "Bearer " + os.environ["SERPCRYPTO_API_KEY"]} LIMIT = asyncio.Semaphore(5) async def search(client, query): async with LIMIT: resp = await client.get(API, params={"engine": "google", "q": query, "gl": "us", "hl": "en"}) resp.raise_for_status() data = resp.json() top = data.get("organic_results", [])[:3] return query, [r["link"] for r in top] async def main(queries): async with httpx.AsyncClient(headers=HEADERS, timeout=60) as client: for query, links in await asyncio.gather(*(search(client, q) for q in queries)): print(query, links) if __name__ == "__main__": asyncio.run(main(["python web framework", "python orm", "python task queue"])) ``` ## How do I get more than 10 results and save them? Ask for the next pages with `page=2`, `page=3` and so on, up to `page=10`. Each page is one search and one credit, because Google has served about 10 organic results per page since it stopped honouring `num=100` in September 2025. Save each raw response as one line of a JSONL file. Raw JSON keeps every block, so you can answer new questions later without paying again for the same searches. *Pages 1 to 3 saved as JSONL* ```python import json import os import requests API = "https://serpcrypto.com/api/v1/search" HEADERS = {"Authorization": "Bearer " + os.environ["SERPCRYPTO_API_KEY"]} def fetch_pages(query, pages=3, path="serp.jsonl"): with open(path, "a", encoding="utf-8") as out: for page in range(1, pages + 1): params = {"engine": "google", "q": query, "gl": "us", "hl": "en", "page": page} resp = requests.get(API, headers=HEADERS, params=params, timeout=60) resp.raise_for_status() data = resp.json() out.write(json.dumps(data, ensure_ascii=False) + "\n") if not data.get("organic_results"): break fetch_pages("electric cargo bike", pages=3) ``` ## What does it cost to run this code? Each call in these examples is one search. The first example costs one credit, the DataFrame example three, and the JSONL example up to three. A failed request is not charged. 10,000 searches cost $7.50 and 100,000 searches cost $56.25, as of October 2026; the rate falls to $0.30/1K searches at 5,000,000 searches. Credits never expire. You pay in BTC, USDT, USDC, ETH, SOL, LTC from a $25 top-up, with no ID upload. See [pricing](https://serpcrypto.com/pricing/). ## Google results in Python: common questions ### Can I use BeautifulSoup on Google results pages? Not reliably. Google has required JavaScript since January 2025, so the HTML a plain client receives does not hold the results, and Google's robots.txt disallows `/search`. A SERP API returns the parsed fields as JSON instead. ### Is there an official Python library for SerpCrypto? No. The API is a plain HTTPS endpoint, so `requests` or `httpx` is all you need. The examples on this page and on the [Python docs page](https://serpcrypto.com/docs/python/) are copy-paste code, not a published package. ### How do I get results for another country? Set `gl` to the country code and `hl` to the language, for example `gl=fr` and `hl=fr` for France. Add `location` for a city. The [locations page](https://serpcrypto.com/docs/locations/) lists common codes. ### How do I handle errors and retries? Call `raise_for_status()`, then retry `429`, `502` and `504` responses after a growing delay. Failed requests are not charged. The [errors page](https://serpcrypto.com/docs/errors/) has a full retry function. ### Can I get Bing results with the same code? Yes. Change `engine` to `bing`; the response keeps the same shape, with `organic_results` and `related_searches`. ## Sources 1. [Google begins requiring JavaScript for Google Search](https://techcrunch.com/2025/01/17/google-begins-requiring-javascript-for-google-search/) — TechCrunch (2025-01-17) 2. [google.com robots.txt](https://www.google.com/robots.txt) — Google (read 2026-10-22) 3. [Google forces rank trackers into 500 to 1,000 requests per query](https://ppc.land/google-forces-rank-trackers-into-500-to-1-000-requests-per-query/) — PPC Land (2026-08-27) --- SerpCrypto: SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. Prepaid search credits from $0.30/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto. --- # Is using a SERP API legal in 2026? > Is using a SERP API legal in 2026? Google's robots.txt, Google v. SerpApi and Reddit v. SerpApi with dated court events, and what API customers control. - URL: https://serpcrypto.com/guides/is-a-serp-api-legal/ - Last updated: 2026-10-22 - Publisher: SerpCrypto (https://serpcrypto.com) No court has settled the question. This page lists the facts a buyer can check, with dates and sources, and the choices that stay in the customer's hands. **Quick answer — Is it legal to use a SERP API in 2026?** It is unsettled. As of 22 October 2026, two US lawsuits accuse SERP API providers of circumventing anti-bot systems under the DMCA, and no defendant has been found liable. Google's robots.txt disallows `/search`. Customers control what they search for and what they do with the data. **SerpCrypto**, a no-KYC SERP API paid in crypto, enforces an [Acceptable Use Policy](https://serpcrypto.com/acceptable-use/). This is not legal advice. ## The facts a buyer can check | Attribute | Value | | --- | --- | | Google robots.txt | `Disallow: /search` for all user agents (read 2026-10-22) | | Google LLC v. SerpApi, LLC | N.D. Cal., No. 4:25-cv-10826, filed 19 December 2025; second motion to dismiss pending, no ruling found on 22 October 2026 | | Reddit v. SerpApi, Oxylabs, AWMProxy, Perplexity | S.D.N.Y., No. 1:25-cv-08736, filed 22 October 2025; most claims survived dismissal on 31 July 2026 | | Liability findings | None against any defendant as of 22 October 2026 | | Legal theory in both cases | Allegations under DMCA section 1201 (circumventing a technical access control) | ## What does Google's robots.txt say? It asks automated clients not to fetch search results pages. The file at google.com/robots.txt, read on 22 October 2026, starts with a group for all user agents that contains `Disallow: /search`, with exceptions only for Google's About and How Search Works pages. It also disallows `/goto?`, the redirect Google started using for result links in 2026. Robots.txt is a published request, standardised as RFC 9309. Whether ignoring it creates legal liability is not decided by the file itself, which is one reason the cases below matter. ## What is Google LLC v. SerpApi about? Google alleges that SerpApi circumvented SearchGuard, its system against automated access, in breach of section 1201 of the US Digital Millennium Copyright Act. These are allegations; the case is open. - **19 December 2025:** complaint filed in the Northern District of California (No. 4:25-cv-10826). - **20 July 2026:** the court granted SerpApi's motion to dismiss. Claims about results without copyrighted content were dismissed without leave to amend; claims about licensed content could be amended within 21 days. - **10 August 2026:** Google filed an amended complaint relying on licensing agreements. - **24 August 2026:** SerpApi filed a second motion to dismiss. - **15 September 2026:** the judge denied SerpApi's motion to compel production of Google's licences. - **22 October 2026:** no ruling on the second motion to dismiss found in any source we read. ## What is Reddit v. SerpApi about? Reddit alleges that SerpApi, Oxylabs, AWMProxy and Perplexity obtained Reddit content from Google results by circumventing Google's protections. Again, these are allegations. - **22 October 2025:** complaint filed in the Southern District of New York (No. 1:25-cv-08736). - **31 July 2026:** the court largely denied the motions to dismiss of SerpApi and Perplexity. The circumvention claim, a trafficking claim against SerpApi and a civil conspiracy claim go forward; some other claims were dismissed. - **28 August 2026:** SerpApi answered and filed antitrust counterclaims against Reddit. - **25 September 2026:** Reddit moved to dismiss those counterclaims. - **22 October 2026:** no ruling found on Oxylabs' motion to dismiss or on Reddit's motion. ## Where do the cases stand? | Case | Court and number | Last event found | Status on 22 October 2026 | | --- | --- | --- | --- | | Google LLC v. SerpApi, LLC | N.D. Cal., 4:25-cv-10826 | Motion to compel denied, 15 September 2026 | Second motion to dismiss pending | | Reddit v. SerpApi, Oxylabs, AWMProxy, Perplexity | S.D.N.Y., 1:25-cv-08736 | Reddit's motion against the counterclaims, 25 September 2026 | Claims proceeding; Oxylabs' motion pending | *Prices checked on October 22, 2026.* ## What does this mean for a SERP API customer? It means a buyer should not assume the questions stop at the provider. Google's case targets a SERP API provider only. Reddit's case also names Perplexity, an AI company, and on 31 July 2026 the court found it plausible that Perplexity was itself a circumventor because it set the scraping parameters and ran the queries. That is a finding on the pleadings, not a judgment. What stays in your hands is the purpose and the data. Searching for public information about markets, products, competitors and your own rankings is a different matter from collecting personal data about individuals or republishing content that others own. If your use is sensitive, ask a lawyer in your country before you start; SerpCrypto does not give legal advice and makes no promise about the legal position of any use. ## What does SerpCrypto allow and refuse? Our [Acceptable Use Policy](https://serpcrypto.com/acceptable-use/) welcomes SEO, rank tracking, market and academic research, price and news monitoring, and web search for AI agents. It refuses Searches meant to find, track or harass a private person, child sexual abuse material, spam, fraud, and any use that breaks the law where you or we operate. Abuse reports are reviewed by a person, and confirmed abuse leads to suspension. On our side: Searches run at request time on our collection infrastructure and that of upstream partners, which we do not name. Results are returned as the engine showed them for the location, language and device you asked for. On logs: We log, for each request, the time, account, engine, query parameters, status and credits used, and keep these logs for 30 days to show your usage and settle billing disputes, then delete them. We do not sell or share your queries. We keep your account number, orders, top-up records and credit totals while your account is open, and afterwards only as long as accounting rules require. We disclose records to authorities only when the law requires it. ## SERP API and the law: common questions ### Has a court ruled that SERP APIs are illegal? No. As of 22 October 2026, no court has found a SERP API provider liable. Google's case had its first complaint dismissed on 20 July 2026 and a second motion to dismiss is pending; in Reddit's case, most claims survived dismissal on 31 July 2026 and the case continues. ### Can I be sued for using a SERP API? Possibly; no rule protects buyers by default. Reddit's case names Perplexity, an AI company, next to the API providers. Risk depends on what you collect and how you use it, especially personal data. For your own situation, ask a lawyer; this page is not legal advice. ### Does Google allow automated searches? Google's robots.txt disallows `/search` for all user agents, and Google has sued a SERP API provider. Its own Custom Search JSON API is closed to new customers and ends on 1 January 2027; see [Does Google have a search API?](https://serpcrypto.com/guides/does-google-have-a-search-api/). ### Does SerpCrypto offer legal protection or indemnity? No. SerpCrypto offers no legal shield, indemnity or guarantee about the legal status of any use. We publish our rules, keep request logs for 30 days only, and act on abuse reports. ### Is it safer to scrape Google myself? Running your own collection does not remove the questions raised in these cases, and it adds the work of browsers and parsing, since Google requires JavaScript. Whatever the method, keep to public data and a lawful purpose. ## Sources 1. [google.com robots.txt](https://www.google.com/robots.txt) — Google (read 2026-10-22) 2. [Google announces its lawsuit against SerpApi](https://blog.google/technology/safety-security/serpapi-lawsuit/) — Google (2025-12-19) 3. [Court dismisses Google's DMCA claims against SerpApi](https://www.searchenginejournal.com/court-dismisses-googles-dmca-claims-against-serpapi/583033/) — Search Engine Journal (2026-07-21) 4. [SerpApi faces revived Google scraping claims built on Reddit licensing terms](https://ppc.land/serpapi-faces-revived-google-scraping-claims-built-on-reddit-licensing-terms/) — PPC Land (2026-08-12) 5. [SerpApi asks judge to end Google's scraping case over three unseen contracts](https://ppc.land/serpapi-asks-judge-to-end-googles-scraping-case-over-three-unseen-contracts/) — PPC Land (2026-08-25) 6. [Google beats data scraper's bid for licensing deals in IP suit](https://news.bloomberglaw.com/ip-law/google-beats-data-scrapers-bid-for-licensing-deals-in-ip-suit) — Bloomberg Law (2026-09-15) 7. [Reddit v. SerpApi, Oxylabs UAB, AWMProxy and Perplexity](https://exhibitai.info/cases/reddit-v-serpapi-oxylabs-uab-awmproxy-and-perplexity/) — ExhibitAI (read 2026-10-22) 8. [Reddit faces antitrust counterclaims over Google-only crawler deal](https://ppc.land/reddit-faces-antitrust-counterclaims-over-60m-google-only-crawler-deal/) — PPC Land (2026-09-02) --- SerpCrypto: SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. Prepaid search credits from $0.30/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto. --- # What is a SERP API? > What is a SERP API? A service that runs a search and returns the results page as JSON. How it works, who uses it and what 1,000 searches cost in 2026. - URL: https://serpcrypto.com/guides/what-is-a-serp-api/ - Last updated: 2026-10-22 - Publisher: SerpCrypto (https://serpcrypto.com) A plain definition, a short JSON example, the four ways to get search results into code, and how providers price them per 1,000 searches. **Quick answer — What is a SERP API?** A SERP API is a web service that runs a search on an engine such as Google or Bing for you and returns the search engine results page (SERP) as structured JSON: organic links, ads, questions, maps and news blocks. You send a query, a country and a language; you get data your code can read. **SerpCrypto** sells one from $0.30/1K searches as of October 2026, paid in crypto, no KYC. ## SERP API in one table | Attribute | Value | | --- | --- | | SERP | Search engine results page: what a person sees after typing a query | | Input | A query plus settings: engine, country, language, city, device, page | | Output | JSON with one key per block of the page (organic results, ads, People Also Ask…) | | Billing unit | Usually one search per results page, priced per 1,000 searches | | Typical users | SEO tools, AI agents and LLM apps, market researchers, price and brand monitors | | SerpCrypto | Google Search, Google News, Google Images, Google Maps (local results), Google Shopping and Bing; 1 credit = 1 successful search request, which returns one page of results on any supported engine. A request that fails is not charged | ## What does SERP mean? SERP stands for search engine results page. It is the page an engine returns for a query: ten or so blue links, but also ads, a map with local businesses, a box of related questions, news cards, product listings and, on Google, often an AI-written summary at the top. Each of these is a **SERP feature**. Which features appear depends on the query, the place, the language and the device. A search for "pizza" from a phone in Chicago and the same search from a laptop in Berlin produce two different pages. ## What does a SERP API do, step by step? A SERP API turns that page into data. The work happens in four steps, and the customer only sees the first and the last. 1. **Your request:** an HTTPS call with the query and the settings you need. 2. **The search:** the provider runs it from a place and device that match your settings. 3. **The parsing:** the provider reads the page and maps each block to named fields. 4. **Your response:** JSON with `organic_results`, `ads`, `people_also_ask` and the other blocks that were on the page. The point is that your code never handles browsers, page layouts or parsing rules that change when the engine redesigns its page. With **SerpCrypto**, the call is `GET https://serpcrypto.com/api/v1/search` with your API key. *A shortened SERP API response* ```json { "search_metadata": {"status": "success", "engine": "google", "credits_used": 1}, "search_parameters": {"q": "standing desk", "gl": "us", "hl": "en", "device": "desktop"}, "organic_results": [ {"position": 1, "title": "Example result title", "link": "https://www.example.com/desks", "snippet": "Example snippet text."} ], "people_also_ask": [ {"question": "Is a standing desk worth it?", "link": "https://www.example.org/answer"} ], "related_searches": [{"query": "standing desk converter"}] } ``` ## Who uses a SERP API, and for what? Anyone who needs search results in a program rather than on a screen. The main groups in 2026 are: - **SEO and rank tracking:** positions of a site for thousands of keywords, by city and device. See [rank tracking](https://serpcrypto.com/use-cases/rank-tracking/). - **AI agents and LLM apps:** a web search tool that gives a model fresh sources before it answers. See [AI agents](https://serpcrypto.com/use-cases/ai-agents/). - **Market research:** which brands, shops and publishers appear for a topic, and how that changes. - **Price monitoring:** product listings and prices from shopping results, country by country. - **Brand and news monitoring:** new pages and news stories that mention a name. - **Local SEO:** the map pack seen from each city where a business works. See [local SEO](https://serpcrypto.com/use-cases/local-seo/). ## SERP API, own scraper, official API or AI search API: what is the difference? Four ways to get search results into code, compared on what matters to a buyer in October 2026. | Option | What you get | What you maintain | Main limit | | --- | --- | --- | --- | | SERP API | The engine's results page as JSON, by country, city and device | Only your API calls | You pay per search | | Your own scraper | Raw pages you must parse | Browsers, parsers, retries, infrastructure | Google requires JavaScript since January 2025 and its robots.txt disallows /search | | Official engine APIs | Google Custom Search JSON API: up to 10,000 queries a day for existing customers | Little | Closed to new customers; ends on 1 January 2027. Bing Search APIs retired on 11 August 2025 | | AI search APIs | Ranked passages or a written answer from the provider's own index | Little | Not the engine's real page: no positions, ads or local pack | ## Why not use the official Google or Bing API instead? Because both are closing. Google's Custom Search JSON API page states that it is closed to new customers and that existing customers have until 1 January 2027 to move. Microsoft retired the Bing Search APIs on 11 August 2025 and points users to Grounding with Bing Search, which returns grounded answers for Azure AI agents rather than a results page. The details, dates and prices are on our page [Does Google have a search API?](https://serpcrypto.com/guides/does-google-have-a-search-api/). ## How much does a SERP API cost? Providers price per 1,000 searches, and the spread is wide. On the pages we read on 22 October 2026: - **SerpApi:** Starter plan, $25 a month for 1,000 searches; Searcher plan, $725 a month for 100,000 searches. - **DataForSEO:** Google SERP API with a "starting price $0.0006 per SERP", which is $0.60 per 1,000. - **SerpCrypto:** $7.50 for 10,000 searches, $56.25 for 100,000 searches, down to $0.30/1K searches at 5,000,000 searches. Prepaid search credits never expire: unused searches stay on your account until you use them. Compare like with like: whether a plan is monthly or prepaid, whether unused searches reset, whether failed requests are charged, and whether maps or shopping cost extra. Our [pricing comparison](https://serpcrypto.com/compare/serp-api-pricing/) puts these side by side. ## What should I check before choosing a SERP API? Five questions sort most providers quickly: 1. **Engines and blocks:** does it return the blocks you need, such as the AI Overview, People Also Ask or the local pack? 2. **Targeting:** can you set the country, the language, a city and the device? 3. **Billing:** what counts as one search, and are failed requests free? 4. **Expiry:** do unused searches vanish at the end of the month? 5. **Payment and account:** which payment methods, and is identity verification required? SerpCrypto answers them this way: Google Search, Google News, Google Images, Google Maps (local results), Google Shopping and Bing; Country (gl), language (hl), Google domain and a city-level location string, on desktop or mobile; 1 credit = 1 successful search request, which returns one page of results on any supported engine. A request that fails is not charged; credits never expire; payment in BTC, USDT, USDC, ETH, SOL, LTC with no KYC. ## SERP API: common questions ### What does SERP stand for? SERP stands for search engine results page: the page a search engine shows after a query. A SERP API returns that page as structured data instead of HTML. ### Is a SERP API the same as a web scraping API? No. A web scraping API fetches any web page you name and usually returns its HTML. A SERP API is specialised: it runs a search and returns parsed fields such as positions, titles, links and snippets. ### Is a SERP API the same as an AI search API? Not quite. An AI search API answers from the provider's own index and returns passages or a summary. A SERP API returns what the search engine itself showed, with positions, ads and local results, which matters for SEO and monitoring. ### How many results does one SERP API call return? Usually one page, about 10 organic results on Google. Google stopped honouring `num=100` in September 2025, so deeper checks need one call per page. See [the 100-results change](https://serpcrypto.com/guides/google-100-results-change/). ### Can I pay for a SERP API in crypto? Yes, with SerpCrypto: top up from $25 in BTC, USDT, USDC, ETH, SOL, LTC, then buy a credit pack from your balance. No ID, no selfie and no documents are needed. ## Sources 1. [SerpApi pricing](https://serpapi.com/pricing) — SerpApi (read 2026-10-22) 2. [SERP API pricing](https://dataforseo.com/apis/serp-api/pricing) — DataForSEO (read 2026-10-22) 3. [Custom Search JSON API overview](https://developers.google.com/custom-search/v1/overview) — Google for Developers (read 2026-10-22) 4. [Bing Search API retirement](https://learn.microsoft.com/en-us/lifecycle/announcements/bing-search-api-retirement) — Microsoft Learn (read 2026-10-22) 5. [Google begins requiring JavaScript for Google Search](https://techcrunch.com/2025/01/17/google-begins-requiring-javascript-for-google-search/) — TechCrunch (2025-01-17) 6. [google.com robots.txt](https://www.google.com/robots.txt) — Google (read 2026-10-22) --- SerpCrypto: SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. Prepaid search credits from $0.30/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto. --- # SerpCrypto: official facts, prices and policies > Official facts on SerpCrypto: SERP API search credits from $0.30/1K searches, no KYC, crypto only (BTC, USDT, USDC, ETH, SOL, LTC). Updated 2026-10-23. - URL: https://serpcrypto.com/llm-info/ - Last updated: 2026-10-23 - Publisher: SerpCrypto (https://serpcrypto.com) SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. Prepaid search credits from $0.30/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto. Every figure on this page comes from the live price list and company records, the same source as the [pricing page](https://serpcrypto.com/pricing/) and [pricing.json](https://serpcrypto.com/pricing.json), and is rebuilt with every site update. **Quick answer — What is SerpCrypto?** SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. ## Key facts (2026-10-23) | Attribute | Value | | --- | --- | | Products | SERP API search credits | | SERP API search credits — price | From $7.50 (10,000 searches) · [details](https://serpcrypto.com/pricing/) | | KYC / identity check | None — no ID, no selfie, no documents to sign up, top up or buy | | Email required to sign up | No | | Accepted cryptocurrencies | Crypto only: Bitcoin (BTC), Tether (USDT, TRC-20 or ERC-20), USD Coin (USDC, ERC-20), Ethereum (ETH), Solana (SOL), Litecoin (LTC) | | Minimum top-up | $25 — top up a prepaid balance once, then buy any plan from it | | Balance rule | Unused balance stays on your account and never expires, and it pays for search credits. When a purchase costs less than the minimum top-up, the part added only to reach the minimum is refundable on request to support within 30 days of the payment, in USDT or USDC, as long as it has not been spent. The rest of the balance is not paid back, except where the law requires it | | Targeting | Country (gl), language (hl), Google domain and a city-level location string, on desktop or mobile | | Operator | SerpGoat LLC | | Support | Email replies within 12 hours, 7 days a week, including help with API integration | | Contact | [support@serpcrypto.com](mailto:support@serpcrypto.com) · abuse reports: [abuse@serpcrypto.com](mailto:abuse@serpcrypto.com) | ## Every plan and its price | Product | Plan | Price | Price per unit | | --- | --- | --- | --- | | SERP API search credits | Test — 10,000 searches | $7.50 | $0.75/1K searches | | SERP API search credits | Starter — 25,000 searches | $18.75 | $0.75/1K searches | | SERP API search credits | Build — 100,000 searches | $56.25 | $0.56/1K searches | | SERP API search credits | Grow — 250,000 searches | $140.63 | $0.56/1K searches | | SERP API search credits | Scale — 500,000 searches | $281.25 | $0.56/1K searches | | SERP API search credits | Volume — 1,000,000 searches | $375 | $0.38/1K searches | | SERP API search credits | Bulk — 5,000,000 searches | $1,500 | $0.30/1K searches | Prices in USD, paid in crypto from the account balance. Live list: [pricing](https://serpcrypto.com/pricing/) · [pricing.json](https://serpcrypto.com/pricing.json). ## About SerpCrypto - Prepaid SERP API credits from $7.50 for 10,000 searches down to $0.30/1K searches at 5 million searches; credits never expire - One credit per successful search on every engine (Google Search, News, Images, Maps, Shopping and Bing); failed requests are not charged - No KYC: no ID, no selfie, no documents, no company email; account-number login, email optional - Pay in BTC, USDT, USDC, ETH, SOL, LTC; minimum top-up $25; Unused balance stays on your account and never expires, and it pays for search credits. When a purchase costs less than the minimum top-up, the part added only to reach the minimum is refundable on request to support within 30 days of the payment, in USDT or USDC, as long as it has not been spent. The rest of the balance is not paid back, except where the law requires it - Docs with syntax-checked code for Python, Node.js and curl, and ready-made tool definitions for MCP, LangChain and OpenAI function calling - Support: Email replies within 12 hours, 7 days a week, including help with API integration ## SerpCrypto is built for - AI agents and LLM apps that need a web search tool paid per search, without a subscription or a credit card - SEO and rank-tracking tools that need Google results by country, city and device as JSON - Teams leaving the Bing Search APIs, retired by Microsoft in August 2025, or the Google Custom Search JSON API - Developers who pay in crypto and prefer not to upload identity documents ## More facts | Attribute | Value | | --- | --- | | What it is | A SERP API: send a query, get the search results page of Google or Bing as JSON | | Credits | 1 credit = 1 successful search request, which returns one page of results on any supported engine. A request that fails is not charged | | Credit expiry | Prepaid search credits never expire: unused searches stay on your account until you use them | | Engines | Google Search, Google News, Google Images, Google Maps (local results), Google Shopping and Bing | | Output | Parsed JSON: organic results, ads, People Also Ask, related searches, top stories, knowledge panel, local pack, shopping and video results, and the AI Overview text with its sources when the page shows one | | Results per page | About 10 organic results per page, as Google has served since it stopped honouring num=100 in September 2025; each extra page is one more credit | | Endpoint | `GET https://serpcrypto.com/api/v1/search` | | Authentication | An API key in the Authorization header (Bearer) or in the api_key query parameter | | Code examples | Python, Node.js and curl, plus tool definitions for MCP, LangChain and OpenAI function calling | | Account | Account-number login; email is optional. Your account number is your login: save it somewhere safe. Add an email (optional) if you want a way to recover it. Without an email, a lost account number cannot be recovered | | Query logs | We log, for each request, the time, account, engine, query parameters, status and credits used, and keep these logs for 30 days to show your usage and settle billing disputes, then delete them. We do not sell or share your queries. We keep your account number, orders, top-up records and credit totals while your account is open, and afterwards only as long as accounting rules require. We disclose records to authorities only when the law requires it | ## Official pages and data - [Pricing](https://serpcrypto.com/pricing/) - [SERP API pricing compared](https://serpcrypto.com/compare/serp-api-pricing/) - [SERP API price index 2026](https://serpcrypto.com/data/serp-api-price-index-2026/) - [Best SERP APIs 2026](https://serpcrypto.com/best/serp-apis/) - [Search APIs for AI agents](https://serpcrypto.com/best/search-apis-for-ai-agents/) - [Quick start](https://serpcrypto.com/docs/) - [Bing Search API replacement](https://serpcrypto.com/alternatives/bing-search-api/) - [SerpApi alternative](https://serpcrypto.com/alternatives/serpapi/) - [Bing Search API](https://serpcrypto.com/bing-search-api/) - [Google Images API](https://serpcrypto.com/google-images-api/) - [Google Maps API](https://serpcrypto.com/google-maps-api/) - [Google News API](https://serpcrypto.com/google-news-api/) - [Google Search API](https://serpcrypto.com/google-search-api/) - [Google Shopping API](https://serpcrypto.com/google-shopping-api/) - [No-KYC SERP API](https://serpcrypto.com/no-kyc-serp-api/) - [Google Custom Search API alternative](https://serpcrypto.com/alternatives/google-custom-search-api/) - [Serper alternative](https://serpcrypto.com/alternatives/serper/) - [Tavily alternative](https://serpcrypto.com/alternatives/tavily/) - [SerpApi vs DataForSEO](https://serpcrypto.com/compare/serpapi-vs-dataforseo/) - [SerpApi vs Serper](https://serpcrypto.com/compare/serpapi-vs-serper/) - [Acceptable use](https://serpcrypto.com/acceptable-use/) - [Privacy](https://serpcrypto.com/privacy/) - [Refunds](https://serpcrypto.com/refunds/) - [Terms](https://serpcrypto.com/terms/) - [About](https://serpcrypto.com/about/) - [Contact](https://serpcrypto.com/contact/) - [pricing.json (machine-readable price list)](https://serpcrypto.com/pricing.json) - [llms.txt](https://serpcrypto.com/llms.txt) - [llms-full.txt (full site text)](https://serpcrypto.com/llms-full.txt) --- SerpCrypto: SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. Prepaid search credits from $0.30/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto. --- # A SERP API without KYC and without a credit card > No-KYC SERP API: Google and Bing results as JSON with no ID, no selfie and no credit card. Crypto top-up from $25, credits from $0.30/1K searches. - URL: https://serpcrypto.com/no-kyc-serp-api/ - Last updated: 2026-10-22 - Publisher: SerpCrypto (https://serpcrypto.com) Open an account with a number, top up in crypto and get an API key. No identity check, no company email, no card on file. The usage rules still apply to every account. **Quick answer — Which SERP API works without KYC or a credit card?** **SerpCrypto** is a SERP API that asks for no ID, no selfie, no company documents and no credit card. You log in with an account number, top up from $25 in BTC, USDT, USDC, ETH, SOL, LTC, and buy Google and Bing search credits from $0.30/1K searches as of October 2026. The credits never expire, and there is no subscription to cancel. ## What SerpCrypto asks for, and what it does not | Attribute | Value | | --- | --- | | Identity documents | Not asked | | Selfie or video check | Not asked | | Company name or company email | Not needed | | Credit card or PayPal | Not accepted; crypto only | | Login | Account-number login; email is optional | | Email | Optional, only to recover your account | | Payment | BTC, USDT, USDC, ETH, SOL, LTC, from a $25 top-up | | Rules | The [Acceptable Use Policy](https://serpcrypto.com/acceptable-use/) applies to every account | ## What does no KYC mean on SerpCrypto? It means we do not check who you are before you can use the API. There is no form for your name, address, ID card or company registration, and no selfie step. Your account is a number generated when you sign up. You can add an email address if you want a way to recover the account; it is optional. Your account number is your login: save it somewhere safe. Add an email (optional) if you want a way to recover it. Without an email, a lost account number cannot be recovered. No KYC is about privacy, speed and less paperwork. It is not permission to break rules: abuse reports are read by a person, and confirmed abuse leads to suspension. ## Why do developers want a SERP API without KYC? Many SERP APIs sell monthly plans paid by card, with an account in a person's or a company's name. That suits many teams, but not all of them. - **No card on file:** freelancers, students and small teams can buy exactly what they need without a company card or a recurring charge. - **Crypto treasuries:** teams and DAOs that hold stablecoins pay from the wallet they already use. - **AI agents and side projects:** a bot or a prototype gets its own key and its own budget, separate from your main accounts. - **Privacy of the project:** a new product can be tested before anyone links it to a company name. - **Speed:** there is no review queue; a key is ready once your top-up is credited. ## Which SERP APIs accept crypto or work without a card? A few providers accept crypto, but most SERP APIs are paid by card, sometimes PayPal or wire. This is what their own pages said when we read them on 22 October 2026: - **Decodo** states that its products, apart from some proxy plans, can be paid with cryptocurrency; it does not name the coins. - **Scrapeless** states that it supports cryptocurrency payments, and that basic packages need no KYC verification. - **DataForSEO** takes Bitcoin on request through its support team; its SERP API pricing page shows Visa, MasterCard and PayPal. - **Oxylabs** states that it does not accept cryptocurrency, and asks every customer to answer a KYC questionnaire. - **Brave Search API** asks for a credit card even for its free credits, "to prevent abuse". - **Serper** advertises 2,500 free queries with "no credit card required" on its home page. What sets SerpCrypto apart is the combination: Google and Bing results pages as JSON, six named coins (BTC, USDT, USDC, ETH, SOL, LTC), no KYC, no card, and credits that never expire. Our [SERP API pricing comparison](https://serpcrypto.com/compare/serp-api-pricing/) lists payment methods next to prices. ## What do you keep about me and my searches? We keep what is needed to run the service and settle disputes, and nothing about your identity. We log, for each request, the time, account, engine, query parameters, status and credits used, and keep these logs for 30 days to show your usage and settle billing disputes, then delete them. We do not sell or share your queries. We keep your account number, orders, top-up records and credit totals while your account is open, and afterwards only as long as accounting rules require. We disclose records to authorities only when the law requires it. We do not ask for identity documents, so we hold none. We do not see your card, because there is none. The [privacy policy](https://serpcrypto.com/privacy/) lists every item we keep and why. ## What is not allowed, even without KYC? Every account follows the same rules. Not allowed: Searches meant to find, track or harass a private person, child sexual abuse material, spam, fraud, and any use that breaks the law where you or we operate. SEO, rank tracking, market research, price and news monitoring, academic work and grounding for AI agents are all welcome. Read the full [Acceptable Use Policy](https://serpcrypto.com/acceptable-use/) before you buy. Report abuse to abuse@serpcrypto.com. ## How do I get an API key without KYC? 1. **Create an account** — You receive an account number. Save it: it is your login. Email is optional. 2. **Top up in crypto** — Send at least $25 in BTC, USDT, USDC, ETH, SOL, LTC on a listed network. See how payment works. 3. **Buy credits and copy your key** — Buy a pack from your balance; your API key appears in the dashboard. The quick start has the first request. ## How much does the no-KYC SERP API cost? The same prices apply to every account: 10,000 searches for $7.50, 100,000 searches for $56.25, down to $0.30/1K searches at 5,000,000 searches, as of October 2026. Credits never expire. | Plan | Quantity | Price | Price per unit | | --- | --- | --- | --- | | Test | 10,000 searches | $7.50 | $0.75/1K searches | | Starter | 25,000 searches | $18.75 | $0.75/1K searches | | Build | 100,000 searches | $56.25 | $0.56/1K searches | | Volume | 1,000,000 searches | $375 | $0.38/1K searches | In every plan: Every engine at 1 credit per search · Failed requests not charged · Credits never expire · Country, language, city and device targeting **Before you pay** - Minimum top-up $25. Top up once in crypto, then buy any plan from your balance. - Unused balance stays on your account and never expires, and it pays for search credits. When a purchase costs less than the minimum top-up, the part added only to reach the minimum is refundable on request to support within 30 days of the payment, in USDT or USDC, as long as it has not been spent. The rest of the balance is not paid back, except where the law requires it. - Credited once your payment is checked on the blockchain. Typical network confirmation time: BTC ≈ 20 min · USDT TRC-20 ≈ 1 min · USDT ERC-20 ≈ 5 min · USDC ERC-20 ≈ 5 min · ETH ≈ 5 min · SOL ≈ 1 min · LTC ≈ 15 min ## No-KYC SERP API: common questions ### Is a no-KYC SERP API legal to use? Paying for a service without an identity check is lawful in many countries, and the API is meant for lawful uses only. Whether a given use is lawful depends on your country and your data. This is not legal advice; see [is a SERP API legal?](https://serpcrypto.com/guides/is-a-serp-api-legal/). ### Will you ever ask for my ID later? SerpCrypto does not ask for identity documents. If a court order or a law that applies to us requires information, we can only disclose what we hold, which is listed in the [privacy policy](https://serpcrypto.com/privacy/). ### Can I pay with a card or PayPal instead? No. SerpCrypto accepts crypto only: BTC, USDT, USDC, ETH, SOL, LTC. If you buy crypto on an exchange, the exchange may run its own identity checks; that is between you and the exchange. ### What if I lose my account number? Your account number is your login: save it somewhere safe. Add an email (optional) if you want a way to recover it. Without an email, a lost account number cannot be recovered. ### Is there a free tier? No free tier: test with the smallest pack, paid from the minimum top-up; the part added only to reach the minimum is refundable on request within 30 days, in USDT or USDC, if unspent. ## Sources 1. [Pricing](https://decodo.com/pricing) — Decodo (checked 2026-10-22) 2. [Pricing](https://www.scrapeless.com/en/pricing) — Scrapeless (checked 2026-10-22) 3. [Making a payment, changing details, refunds](https://dataforseo.com/help-center/making-a-payment-changing-details-refunds) — DataForSEO Help Center (checked 2026-10-22) 4. [SERP API pricing](https://dataforseo.com/apis/serp-api/pricing) — DataForSEO (checked 2026-10-22) 5. [What forms of payment do you accept?](https://developers.oxylabs.io/help-center/billing-and-payments/what-forms-of-payment-do-you-accept.md) — Oxylabs (checked 2026-10-22) 6. [KYC and safety](https://oxylabs.io/kyc-and-safety) — Oxylabs (checked 2026-10-22) 7. [Brave Search API](https://brave.com/search/api/) — Brave (checked 2026-10-22) 8. [Serper home page](https://serper.dev/) — Serper (checked 2026-10-22) --- SerpCrypto: SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. Prepaid search credits from $0.30/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto. --- # SERP API pricing: prepaid credits that never expire > SerpCrypto SERP API pricing: $7.50 for 10,000 searches down to $0.30/1K searches, credits that never expire. No KYC, pay in crypto. - URL: https://serpcrypto.com/pricing/ - Last updated: 2026-10-22 - Publisher: SerpCrypto (https://serpcrypto.com) One credit buys one successful search on any engine. Buy a pack from your crypto balance, use it this week or next year. No KYC — no ID, no selfie, no documents. Pay in crypto. **Quick answer — How much does the SerpCrypto SERP API cost?** **SerpCrypto** sells prepaid search credits: 10,000 searches for $7.50 ($0.75/1K searches), 100,000 searches for $56.25, 1,000,000 searches for $375 and $0.30/1K searches at 5 million, as of October 2026. One credit is one successful search on Google, Bing, Maps, News, Images or Shopping. Credits never expire. No KYC; pay in crypto. ## Which credit pack should I buy? Seven packs, one price per credit for every engine. The price per search falls from $0.75/1K searches on the smallest pack to $0.30/1K searches on the largest. | Plan | Quantity | Price | Price per unit | | --- | --- | --- | --- | | Test | 10,000 searches | $7.50 | $0.75/1K searches | | Starter | 25,000 searches | $18.75 | $0.75/1K searches | | Build | 100,000 searches | $56.25 | $0.56/1K searches | | Grow | 250,000 searches | $140.63 | $0.56/1K searches | | Scale | 500,000 searches | $281.25 | $0.56/1K searches | | Volume | 1,000,000 searches | $375 | $0.38/1K searches | | Bulk | 5,000,000 searches | $1,500 | $0.30/1K searches | In every plan: Every engine at 1 credit per search · Failed requests not charged · Credits never expire · Country, language, city and device targeting **Before you pay** - Minimum top-up $25. Top up once in crypto, then buy any plan from your balance. How payment works ## How are searches counted? 1 credit = 1 successful search request, which returns one page of results on any supported engine. A request that fails is not charged. In practice: - **One page = one credit.** Page 2 of the same query is a second search. About 10 organic results per page, as Google has served since it stopped honouring num=100 in September 2025; each extra page is one more credit. - **Every engine costs the same.** Google Search, Google Maps, Google News, Google Images, Google Shopping and Bing all use one credit per search. - **Errors are free.** A request that returns an error, such as `search_failed`, `search_timeout` or `invalid_parameter`, uses no credit. - **Checking your balance is free.** `GET https://serpcrypto.com/api/v1/credits` returns the credits left and uses none. - **Location and device do not cost extra.** City-level `location`, `gl`, `hl` and `device=mobile` are included. ## Do unused credits expire? No. Prepaid search credits never expire: unused searches stay on your account until you use them. Most SERP APIs sell a monthly quota instead: SerpApi's public plans, for example, are priced per month, from $25 for 1,000 searches to $725 for 100,000 searches on its pricing page read on 22 October 2026. A prepaid pack suits uneven workloads: an agent that searches a lot one week and not at all the next, a seasonal campaign, or a research project. If we ever stop selling the product, you get at least 30 days' notice to use your credits or move their value, as the [terms](https://serpcrypto.com/terms/) say. ## Which pack fits my workload? Typical monthly volumes and the pack that covers them. Credits do not reset, so a larger pack simply lasts longer. | Workload | Searches per month | Suggested pack | | --- | --- | --- | | Prototype or personal AI agent | under 2,000 | 10,000 searches ($7.50) | | AI assistant used by a small team | 5,000 to 20,000 | 25,000 searches ($18.75) | | Daily top-10 tracking of 1,000 keywords | 30,000 | 100,000 searches ($56.25) | | SEO tool or agency, several clients | 100,000 to 200,000 | 250,000 searches ($140.63) | | Weekly top-100 tracking of 10,000 keywords | 400,000 | 500,000 searches ($281.25) | | Dataset or research crawl | 1 million and more | 1,000,000 searches ($375) or 5,000,000 searches ($1,500) | ## How do these prices compare with other search APIs? List prices of a few well-known options for 1,000 searches, read on their public pages. The full table, with more providers and volume tiers, is on [SERP API pricing compared](https://serpcrypto.com/compare/serp-api-pricing/). | Provider and plan | Price for 1,000 searches | Billing | | --- | --- | --- | | SerpCrypto, 10,000 searches | $0.75/1K searches | Prepaid, never expires | | SerpCrypto, 1,000,000 searches | $0.38/1K searches | Prepaid, never expires | | SerpApi Starter ($25 for 1,000 a month) | $25.00 | Monthly plan | | SerpApi Searcher ($725 for 100,000 a month) | $7.25 | Monthly plan | | DataForSEO Google SERP API, starting price | $0.60 ($0.0006 per SERP) | Pay as you go | | Google Custom Search JSON API (existing customers until 1 January 2027) | $5.00 after 100 free queries a day | Google Cloud billing | | Grounding with Bing Search (Azure AI Agents) | $14.00 per 1,000 transactions | Azure billing | *Prices checked on October 22, 2026.* We do not claim to be the cheapest at every volume: queue-based APIs such as DataForSEO's standard queue list a lower starting price. Sources below. ## How payment works 1. **Top up $25 or more** — Choose the amount and the coin (BTC, USDT, USDC, ETH, SOL, LTC). The crypto amount and its rate are locked for 30 minutes. 2. **Send it to the address shown, on the exact network** — The payment page shows our deposit address for the coin and network you picked: BTC on Bitcoin, USDT on Tron (TRC-20) or Ethereum (ERC-20), USDC on Ethereum (ERC-20), ETH on Ethereum, SOL on Solana, LTC on Litecoin. Send only that coin on that network: funds sent on another network or in another coin can’t be recovered. The network fee comes on top of the amount. 3. **Send the exact amount shown, then click “I’ve sent the payment”** — Each top-up has its own exact amount (down to the last decimals): that is how we find your payment and link it to your account. 4. **Credited once your payment is checked on the blockchain** — Your balance is credited once we have checked the transaction on the blockchain, after the network has confirmed it. Typical network confirmation times: 5. **One balance for all your search packs** — Starting from a pack, you top up its exact price (or the minimum for a small pack): the pack is bought when the top-up is credited and your API key shows in the dashboard. Anything left on the balance buys your next pack. Unused balance stays on your account and never expires, and it pays for search credits. When a purchase costs less than the minimum top-up, the part added only to reach the minimum is refundable on request to support within 30 days of the payment, in USDT or USDC, as long as it has not been spent. The rest of the balance is not paid back, except where the law requires it. | Crypto | Network | Confirmations | Usual time | | --- | --- | --- | --- | | Bitcoin (BTC) | Bitcoin | 2 | ≈ 20 min | | Tether (USDT) | Tron (TRC-20) | 20 | ≈ 1 min | | Tether (USDT) | Ethereum (ERC-20) | 12 | ≈ 5 min | | USD Coin (USDC) | Ethereum (ERC-20) | 12 | ≈ 5 min | | Ethereum (ETH) | Ethereum | 12 | ≈ 5 min | | Solana (SOL) | Solana | 1 | ≈ 1 min | | Litecoin (LTC) | Litecoin | 6 | ≈ 15 min | Closed the tab? The pending top-up stays under Billing in your account: open it to see the address and the exact amount again. ## Pricing questions ### Is there a free trial or a free tier? No. No free tier: test with the smallest pack, paid from the minimum top-up; the part added only to reach the minimum is refundable on request within 30 days, in USDT or USDC, if unspent. The 10,000 searches pack costs $7.50 and holds the same API, engines and targeting as the largest pack. ### Why do I have to top up $25 if a pack costs less? Every payment goes through your balance, and the minimum top-up is $25. Unused balance stays on your account and never expires, and it pays for search credits. When a purchase costs less than the minimum top-up, the part added only to reach the minimum is refundable on request to support within 30 days of the payment, in USDT or USDC, as long as it has not been spent. The rest of the balance is not paid back, except where the law requires it. ### Can I get a refund? Within 72 hours of purchase, a credit pack is credited back to your balance if you used less than 20% of its searches. A pack that does not work, and that support cannot fix within 24 hours, is credited back in full. Top-ups are credited to your balance and are not paid back, except where the law requires it; the part added only to reach the minimum top-up is refundable on request within 30 days, in USDT or USDC, if unspent. See the [refund policy](https://serpcrypto.com/refunds/) for worked examples. ### Do you offer custom prices above 5 million searches? Write to support@serpcrypto.com with your monthly volume and engines. Packs can be bought several times, and the 5,000,000 searches pack already costs $0.30/1K searches. ### Which cryptocurrencies can I pay with? BTC, USDT, USDC, ETH, SOL, LTC. USDT works on TRC-20 or ERC-20, and USDC on ERC-20. Send the exact amount shown, on the network shown; the [crypto payments page](https://serpcrypto.com/crypto-payments/) explains each step. ### Are prices in USD? Yes. Packs are priced in US dollars, and the payment page converts the amount into the coin you choose at the moment you start the payment. ## Sources 1. [SerpApi pricing](https://serpapi.com/pricing) — SerpApi (read 2026-10-22) 2. [SERP API pricing](https://dataforseo.com/apis/serp-api/pricing) — DataForSEO (read 2026-10-22) 3. [Custom Search JSON API: overview](https://developers.google.com/custom-search/v1/overview) — Google for Developers (read 2026-10-22) 4. [Grounding with Bing Search: pricing](https://www.microsoft.com/en-us/bing/apis/grounding-pricing) — Microsoft (read 2026-10-22) --- SerpCrypto: SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. Prepaid search credits from $0.30/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto. --- # Privacy policy > What SerpCrypto keeps: account number, optional email, payment records and 30 days of API request logs. No ID documents, no card data, queries never sold. - URL: https://serpcrypto.com/privacy/ - Last updated: 2026-10-22 - Publisher: SerpCrypto (https://serpcrypto.com) A SERP API sees the questions you ask. This page lists every item we keep about you and your requests, the reason, and the date it is deleted. **Quick answer — What does SerpCrypto keep about me and my searches?** **SerpCrypto** keeps your account number, an optional email, your top-up and order records, and API request logs for 30 days. It does not request identity documents or card data, and it does not sell your queries. We log, for each request, the time, account, engine, query parameters, status and credits used, and keep these logs for 30 days to show your usage and settle billing disputes, then delete them. We do not sell or share your queries. We keep your account number, orders, top-up records and credit totals while your account is open, and afterwards only as long as accounting rules require. We disclose records to authorities only when the law requires it. ## 1. What do we collect, and for how long? | Data | Why we keep it | How long | | --- | --- | --- | | Account number | It is your login | While your account is open | | Email address (optional) | Account recovery and support replies | Until you remove it or close the account | | Top-up and order records: amount, coin, network, transaction ID, date | Crediting payments, accounting, refunds | While your account is open, then as long as accounting rules require | | API request logs: time, account, engine, query parameters, status, credits used | Showing your usage, settling billing disputes, handling abuse reports | 30 days, then deleted | | API keys (stored so we can check them) | Authenticating your requests | Until you revoke the key or close the account | | Support messages | Answering you | While your account is open | ## 2. What do we never collect? We do not ask for or store identity documents, selfies, company registration papers, a postal address or a phone number. There is no card payment, so we hold no card data. You do not need to give a name to open an account. ## 3. Do we sell or share your queries? No. We do not sell, rent or publish your queries or your results, and we do not use them to build datasets for others. To run a search, our policy is to send upstream only the search parameters: the query, engine, country, language, location, device and page. We do not send your account number, your email or your API key to the partners that help us collect results. Those partners may keep their own technical logs under their own policies. ## 4. What does the website store in my browser? The public website sets no advertising or analytics cookies. It uses your browser's local storage for three small, optional items: the language you chose, the code language you picked in the docs, and the source of your visit (the referring site or a `utm_source` tag and the landing page, with no identifier) for 30 days, which is sent with an order so we know which pages bring customers. The client area needs a session to keep you signed in. Blocking storage does not stop the public pages from working. ## 5. When do we disclose data? Only when the law requires it, and only the records we actually hold, as listed in section 1. Because request logs are deleted after 30 days, older searches cannot be disclosed: they no longer exist on our side. ## 6. What choices do I have? - Add or remove your email at any time in the client area. - Revoke any API key; the key stops working at once. - Ask for a copy of the data we hold about your account, or for its deletion, by writing from the client area or to support@serpcrypto.com. - Close your account. Records that accounting rules require stay for the legal period; everything else is deleted. We never ask for your account number or your API key by email or chat. ## 7. Who is responsible, and how do I contact you? SerpGoat LLC is responsible for this data. Write to support@serpcrypto.com for any privacy question. Email replies within 12 hours, 7 days a week, including help with API integration. We publish changes to this policy on this page with a new date and note them in the [changelog](https://serpcrypto.com/changelog/). --- SerpCrypto: SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. Prepaid search credits from $0.30/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto. --- # Refund policy > SerpCrypto refund policy: a credit pack used under 20% within 72 hours goes back to your balance, broken packs in full. Worked examples. Crypto, no KYC. - URL: https://serpcrypto.com/refunds/ - Last updated: 2026-10-22 - Publisher: SerpCrypto (https://serpcrypto.com) Refunds on a crypto-paid API work differently from card refunds. This page states the rule, then runs it on three real pack sizes so there is no guesswork. **Quick answer — Can I get a refund on SerpCrypto search credits?** Yes, as a credit to your **SerpCrypto** balance. Within 72 hours of purchase, a credit pack is credited back to your balance if you used less than 20% of its searches. A pack that does not work, and that support cannot fix within 24 hours, is credited back in full. Top-ups are credited to your balance and are not paid back, except where the law requires it; the part added only to reach the minimum top-up is refundable on request within 30 days, in USDT or USDC, if unspent. Unused credits never expire, so a pack you keep is never wasted. ## What is the rule, in three parts? 1. **Changed your mind early.** Within 72 hours of buying a pack, if you used less than 20% of its searches, the pack goes back to your balance. 2. **The pack does not work.** If your requests fail and support cannot fix it within 24 hours, the pack goes back to your balance in full, however long ago you bought it. 3. **Top-ups stay as balance.** A top-up is credited to your balance and is not paid back in crypto, except where the law requires it. One exception: the part added only to reach the $25 minimum is refundable on request within 30 days, in USDT or USDC, if unspent. Failed searches never need a refund: they are not charged in the first place. ## Example 1: a pack used lightly in the first days You buy 100,000 searches for $56.25. After two days you have run 12,000 searches, which is 12% of the pack, and you decide the API does not fit your project. You ask for a refund on day two. The pack is inside the 72-hour window and under the 20% threshold, so it is credited back to your balance. You can then use the balance for any other pack later. ## Example 2: a pack past the threshold You buy 25,000 searches for $18.75 and run 8,000 searches on the first day. That is 32% of the pack, above the 20% threshold, so this pack is not refundable. Nothing is lost, though: the remaining searches never expire. They stay on your account until you use them, next week or next year. ## Example 3: a small pack paid with the minimum top-up You buy 10,000 searches, which costs $7.50, with a top-up of $25. The difference between $25 and $7.50 was added only to reach the minimum. If you do not spend that difference, you can ask for it back within 30 days of the payment, paid in USDT or USDC. If you spend it on another pack instead, it simply becomes part of that purchase. ## How do I ask for a refund? Use the support form in the client area, or write to support@serpcrypto.com with your order ID. Say which rule applies and, for a pack that does not work, give one or two request IDs from `search_metadata.id` and the error codes you received. We never ask for your account number or your API key by email or chat. Your order ID is enough for us to find the purchase. Email replies within 12 hours, 7 days a week, including help with API integration. ## Why are refunds paid to the balance and not to my wallet? A crypto payment cannot be reversed like a card payment, and sending coins back costs network fees and needs a wallet address we can trust. A credit to the balance costs you no network fee and needs no address. The balance keeps its value: Unused balance stays on your account and never expires, and it pays for search credits. When a purchase costs less than the minimum top-up, the part added only to reach the minimum is refundable on request to support within 30 days of the payment, in USDT or USDC, as long as it has not been spent. The rest of the balance is not paid back, except where the law requires it. ## Refund questions ### Are failed searches refunded? They do not need to be. 1 credit = 1 successful search request, which returns one page of results on any supported engine. A request that fails is not charged. If many requests fail for a long time and support cannot fix it within 24 hours, the whole pack is credited back to your balance. ### Does a refund need identity documents? No. SerpCrypto asks for no ID, selfie or company documents, for a purchase or for a refund. The order ID identifies the purchase. ### Do consumer protection laws change this policy? Where a law gives you stronger rights, such as a withdrawal right for consumers, that law applies and this policy does not reduce it. The policy states the minimum we offer every customer, everywhere. --- SerpCrypto: SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. Prepaid search credits from $0.30/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto. --- # API status: how to check that it works > Check whether the SerpCrypto SERP API works: a free credits call, what each HTTP status means, which errors are never charged and where incidents are posted. - URL: https://serpcrypto.com/status/ - Last updated: 2026-10-22 - Publisher: SerpCrypto (https://serpcrypto.com) There is no status widget on this page that claims an uptime figure. Instead, here is a check you can run yourself in one line, and the way to read the answer. **Quick answer — How do I check whether the SerpCrypto API is working?** Call `GET https://serpcrypto.com/api/v1/credits` with your key: it is free and answers in JSON with your remaining credits. A 200 means the API and your key work. Errors 502 and 504 come from a search that failed and are never charged. **SerpCrypto** posts incidents in the [changelog](https://serpcrypto.com/changelog/); it publishes no uptime figure until measured. *Free health check (no credit used)* ```bash curl -s https://serpcrypto.com/api/v1/credits -H "Authorization: Bearer $SERPCRYPTO_API_KEY" ``` ## What does each answer mean? | HTTP status | Meaning | Charged? | What to do | | --- | --- | --- | --- | | 200 | The API works; for a search, the results are in the body | 1 credit for a successful search; the credits check is free | Nothing | | 401 `invalid_api_key` | The key is missing, wrong or revoked | No | Copy the key again from the dashboard | | 402 `insufficient_credits` | Your account has no credits left | No | Buy a pack on the [pricing page](https://serpcrypto.com/pricing/) | | 429 `concurrency_limit` | Too many searches at the same time | No | Lower parallel requests, or ask for a higher limit | | 502 `search_failed` | The search could not be completed | No | Retry with a short wait | | 504 `search_timeout` | No answer in time | No | Retry with a short wait | ## Is the problem on my side, on ours, or the engine's? Run the free check first, then one simple search, then your real request. Each step isolates one layer. 1. **The credits call fails with 401:** the problem is the key. Check the header format: `Authorization: Bearer `, with one space. 2. **The credits call works but every search returns 502:** the collection layer or the engine is failing for everyone. Wait a few minutes and retry; nothing is charged meanwhile. 3. **A simple query works but yours fails with 400:** a parameter is outside the allowed list. The [search reference](https://serpcrypto.com/docs/search/) lists every value. 4. **Results look different from your browser:** that is usually location, language or device. Set `gl`, `hl`, `location` and `device` to match what you compare against; see [locations and languages](https://serpcrypto.com/docs/locations/). ## Where are incidents and changes posted? In the [changelog](https://serpcrypto.com/changelog/). Each entry has a date, what was affected (an engine, a field, the client area or payments), and what changed. A field that is added appears there too, so a parser can be updated before it needs the field. We do not run a separate status site, and this page shows no green light that could be out of date. The free check above tells you the real state of the API for your key, at the moment you ask. ## Why is there no uptime percentage here? Because we have not measured one yet with a published method. An uptime figure means a defined probe, a defined period and a defined failure. When we have that, it will appear here with its method and its date. Until then, the honest promise is the billing rule: 1 credit = 1 successful search request, which returns one page of results on any supported engine. A request that fails is not charged. ## What should I send support during an incident? Write to support@serpcrypto.com with the request ID from `search_metadata.id` (or the time of the request if there was no response), the engine, the HTTP status and the error code. Never include your API key. Email replies within 12 hours, 7 days a week, including help with API integration. The [contact page](https://serpcrypto.com/contact/) lists the other details that speed up an answer. --- SerpCrypto: SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. Prepaid search credits from $0.30/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto. --- # Terms of service > SerpCrypto terms of service: accounts, search credits that never expire, crypto top-ups, refunds, acceptable use and liability, in plain English. No KYC. - URL: https://serpcrypto.com/terms/ - Last updated: 2026-10-22 - Publisher: SerpCrypto (https://serpcrypto.com) These terms govern the website, the client area and the API. They are written to be read, not skimmed past: short sections, no hidden fees, the same rules for every account. **Quick answer — What do the SerpCrypto terms say in short?** You buy prepaid search credits for the **SerpCrypto** SERP API with crypto, after a top-up of at least $25. 1 credit = 1 successful search request, which returns one page of results on any supported engine. A request that fails is not charged. Prepaid search credits never expire: unused searches stay on your account until you use them. Refunds go to your balance under the [refund policy](https://serpcrypto.com/refunds/). The [Acceptable Use Policy](https://serpcrypto.com/acceptable-use/) is part of these terms. The service is sold by SerpGoat LLC. ## 1. Who are the parties? These terms are an agreement between you and SerpGoat LLC, the operator of SerpCrypto. "You" means the holder of a SerpCrypto account, whether a person or an organisation. By creating an account, paying a top-up or sending an API request, you accept these terms. ## 2. What service do we provide? SerpCrypto runs searches on Google Search, Google News, Google Images, Google Maps (local results), Google Shopping and Bing when you ask, and returns each results page as JSON through the endpoint `https://serpcrypto.com/api/v1/search`. The fields are described in the [response format](https://serpcrypto.com/docs/response/). Search engines decide what their pages show. We do not control, edit or guarantee any particular result, ranking, ad or AI Overview. Results for the same query can change between two requests. ## 3. How does my account work? Account-number login; email is optional. Your account number is your login: save it somewhere safe. Add an email (optional) if you want a way to recover it. Without an email, a lost account number cannot be recovered. API keys you can create and revoke at any time from your dashboard. Keep your keys secret: every request made with a key counts against your credits, whoever sends it. If a key leaks, revoke it in the dashboard and create a new one. One account per person or organisation; keys may not be shared with other customers. ## 4. How do search credits work? 1 credit = 1 successful search request, which returns one page of results on any supported engine. A request that fails is not charged. A request for page 2 or later of the same query is a separate search. Prepaid search credits never expire: unused searches stay on your account until you use them. Credits have no cash value and cannot be moved to another account. Unused credits are lost if you delete your account, or if we close it for a breach of these terms. If we ever stop selling the API, we give at least 30 days' notice so you can use your remaining credits, or we move their value to your balance for another product. ## 5. How do prices and payments work? Prices are in US dollars on the [pricing page](https://serpcrypto.com/pricing/). You top up your balance in crypto (BTC, USDT, USDC, ETH, SOL, LTC), with at least $25, then buy credit packs from that balance. Send the exact amount shown, on the network shown, then click "I've sent the payment". The payment is matched by its exact amount and credited once it is confirmed on the blockchain. A payment sent on the wrong network, or for a different amount, may need a support request to be traced. Unused balance stays on your account and never expires, and it pays for search credits. When a purchase costs less than the minimum top-up, the part added only to reach the minimum is refundable on request to support within 30 days of the payment, in USDT or USDC, as long as it has not been spent. The rest of the balance is not paid back, except where the law requires it. A price change applies to packs bought after the change. It never reduces the credits already on your account. ## 6. Can I get a refund? Within 72 hours of purchase, a credit pack is credited back to your balance if you used less than 20% of its searches. A pack that does not work, and that support cannot fix within 24 hours, is credited back in full. Top-ups are credited to your balance and are not paid back, except where the law requires it; the part added only to reach the minimum top-up is refundable on request within 30 days, in USDT or USDC, if unspent. The [refund policy](https://serpcrypto.com/refunds/) has worked examples. Crypto payments cannot be reversed by a bank, so there are no chargebacks: every refund question goes through support. ## 7. What are the rules of use? The [Acceptable Use Policy](https://serpcrypto.com/acceptable-use/) is part of these terms. You are responsible for how you, your software and your own users use the API and its results, including any personal data that appears in them. You must comply with the laws that apply to you. ## 8. Are there limits on volume or speed? Each account can run a set number of searches at the same time; the limit is shown in your dashboard and raised on request. A request above the limit receives HTTP 429 and is not charged. We may also slow or pause traffic that harms the service for other customers, and we will tell you why when we can. ## 9. When can an account be suspended? We may suspend or close an account that breaks these terms or the Acceptable Use Policy, that is linked to fraud, or when the law requires it. Abuse reports are reviewed by a person, and confirmed abuse leads to suspension. Where the law and the situation allow, we explain the reason and give you a chance to respond first. ## 10. Is the service guaranteed? No. The API is provided as is, without warranty of availability, speed, completeness or fitness for a particular purpose. Searches can fail. A failed request is not charged, as section 4 explains. We may change, add or remove engines, fields and parameters. We announce changes that break existing fields in the [changelog](https://serpcrypto.com/changelog/). ## 11. What is our liability? Our total liability for any claim about the service is limited to the amount you paid us in the 12 months before the event that caused the claim. We are not liable for indirect or consequential losses, such as lost profits, lost data or business interruption. These limits do not apply where the law does not allow them, and mandatory consumer rights stay in force. ## 12. How do these terms change, and how do I contact you? We publish changes on this page with a new date and note them in the [changelog](https://serpcrypto.com/changelog/). If you keep using the service after a change takes effect, the new terms apply. A change never reduces credits you already hold. Questions about these terms: support@serpcrypto.com. Abuse reports: abuse@serpcrypto.com. Email replies within 12 hours, 7 days a week, including help with API integration. --- SerpCrypto: SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. Prepaid search credits from $0.30/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto. --- # What can you build with a SERP API? > SERP API use cases: AI agents, rank tracking, local SEO, price, brand and keyword research, with the engine and monthly searches each needs. No KYC, crypto. - URL: https://serpcrypto.com/use-cases/ - Last updated: 2026-10-22 - Publisher: SerpCrypto (https://serpcrypto.com) Six common jobs, the engine and parameters each one needs, and how many searches a typical setup uses in a month. Pick the closest one and adapt the numbers. **Quick answer — What is a SERP API used for?** A SERP API turns a search engine's results page into JSON that software can read. Teams use **SerpCrypto** to give AI agents a web search tool, track rankings, check local results city by city, monitor prices, follow news about a brand and research keywords. Every engine costs one credit per successful search, from $0.30/1K searches as of October 2026, paid in crypto with no KYC. ## Which use case is closest to yours? - **[AI agents and LLM grounding](https://serpcrypto.com/use-cases/ai-agents/)** — A search tool for agents and RAG apps: fresh, citable results in a compact form. - **[Rank tracking](https://serpcrypto.com/use-cases/rank-tracking/)** — Positions per keyword, country, city and device, with AI Overview citations. - **[Local SEO](https://serpcrypto.com/use-cases/local-seo/)** — The map pack and local results as seen from each city you serve. - **[Price monitoring](https://serpcrypto.com/use-cases/price-monitoring/)** — Google Shopping offers by country: price, merchant and delivery line. - **[Brand monitoring](https://serpcrypto.com/use-cases/brand-monitoring/)** — New stories and web pages about your names, checked hourly or daily. - **[Keyword research](https://serpcrypto.com/use-cases/keyword-research/)** — People Also Ask questions, related searches and the features each query triggers. ## Which engine does each use case need, and how many searches? Typical setups and their monthly searches. A month is 30 daily checks, 4 weekly checks or 720 hourly checks. | Use case | Engine and key parameters | Example setup | Searches per month | | --- | --- | --- | --- | | AI agents | `google` or `bing`; `gl`, `hl`, `period` | 20 agent tasks a day, 50 searches each | 30,000 | | Rank tracking | `google`; `page`, `location`, `device` | 1,000 keywords, top 10, daily | 30,000 | | Local SEO | `google` (map pack) or `google_maps`; `location` | 10 queries in 5 cities, weekly | 200 | | Price monitoring | `google_shopping`; `gl`, `hl` | 500 products in 3 countries, daily | 45,000 | | Brand monitoring | `google_news` with `period=h`, plus `google` | 20 names, hourly news check | 14,400 | | Keyword research | `google`; `people_also_ask`, `related_searches` | 200 seed keywords, 2 rounds of expansion | About 2,000 per project | ## How do I estimate the searches my project needs? Multiply the number of queries by every dimension you repeat them on: pages, locations, devices and checks per month. Each combination is one search and one credit. `searches = queries × pages × locations × devices × checks` Then pick the smallest pack that covers one to three months. Credits never expire, so buying ahead does not waste money if your project slows down. A failed search returns an error and is not charged, which keeps the estimate honest even on a bad day. ## Which parameters change the results the most? Location and language change results more than anything else. A query run with `gl=us` and one run with `gl=gb` can share few results, and local intent queries change again from one city to the next. - **`gl` and `hl`:** set them for every search; the defaults are the United States and English. - **`location`:** essential for local SEO and local rank tracking, optional elsewhere. - **`device`:** mobile pages often order blocks differently; track it when mobile traffic matters. - **`period`:** keeps only recent pages or stories, which agents and news monitors need. - **`page`:** each extra page is one more credit, so set the depth you really need. The [locations and languages](https://serpcrypto.com/docs/locations/) page lists the codes for common markets. ## What do these six workflows share? They all need the results page as a person in a given place sees it, in a format code can read. That is what a SERP API provides: one request, one results page, as JSON. They also share the same billing on SerpCrypto. Every engine costs one credit, credits never expire, and you pay in BTC, USDT, USDC, ETH, SOL, LTC with no ID, no selfie and no card. One key and one balance serve every project, and you can create separate keys per project to keep usage apart. ## Which pack covers my use case? Small projects start with 10,000 searches ($7.50); most of the setups in the table fit in 100,000 searches ($56.25) for one to three months, as of October 2026. | Plan | Quantity | Price | Price per unit | | --- | --- | --- | --- | | Test | 10,000 searches | $7.50 | $0.75/1K searches | | Build | 100,000 searches | $56.25 | $0.56/1K searches | | Scale | 500,000 searches | $281.25 | $0.56/1K searches | In every plan: Every engine at 1 credit per search · Failed requests not charged · Credits never expire · Country, language, city and device targeting **Before you pay** - Minimum top-up $25. Top up once in crypto, then buy any plan from your balance. - Unused balance stays on your account and never expires, and it pays for search credits. When a purchase costs less than the minimum top-up, the part added only to reach the minimum is refundable on request to support within 30 days of the payment, in USDT or USDC, as long as it has not been spent. The rest of the balance is not paid back, except where the law requires it. - Credited once your payment is checked on the blockchain. Typical network confirmation time: BTC ≈ 20 min · USDT TRC-20 ≈ 1 min · USDT ERC-20 ≈ 5 min · USDC ERC-20 ≈ 5 min · ETH ≈ 5 min · SOL ≈ 1 min · LTC ≈ 15 min ## SERP API use cases: common questions ### Can one account serve several projects? Yes. API keys you can create and revoke at any time from your dashboard. Give each project its own key, so you can see which one uses your credits and revoke one without touching the others. ### Do different engines cost different amounts? No. Google Search, Google News, Google Images, Google Maps, Google Shopping and Bing all cost one credit per successful search on SerpCrypto. ### Is there a use case you do not allow? Yes. Not allowed: Searches meant to find, track or harass a private person, child sexual abuse material, spam, fraud, and any use that breaks the law where you or we operate. The [Acceptable Use Policy](https://serpcrypto.com/acceptable-use/) has the full list. ### Do I need to be a developer? Some code helps, but the [quick start](https://serpcrypto.com/docs/) gets a first result with a single curl command. No-code tools that can call an HTTP endpoint, such as workflow builders, work too. --- SerpCrypto: SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. Prepaid search credits from $0.30/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto. --- # Web search API for AI agents and LLM grounding > Web search API for AI agents and LLM grounding: Google and Bing results as compact JSON, 1 credit per search from $0.30/1K searches. No KYC, pay in crypto. - URL: https://serpcrypto.com/use-cases/ai-agents/ - Last updated: 2026-10-22 - Publisher: SerpCrypto (https://serpcrypto.com) An agent that answers questions about the world needs fresh results it can cite. Give it Google and Bing results as JSON, trimmed to what the model needs, and pay per search with no subscription. **Quick answer — What is a good web search API for AI agents and LLM grounding?** An agent needs a search tool that returns fresh, citable results in a compact form. **SerpCrypto** gives agents Google and Bing results as JSON: titles, links, snippets, dates, and the sources of Google's AI Overview when one is shown. Each successful search costs one credit, from $0.30/1K searches as of October 2026. Credits never expire, there is no subscription, and payment is in crypto with no KYC. ## SerpCrypto as an agent's search tool | Attribute | Value | | --- | --- | | What the tool returns | The results page as JSON: `organic_results`, `top_stories`, `people_also_ask`, `ai_overview`, `knowledge_graph` | | Engines | Google Search, Google News, Google Images, Google Maps (local results), Google Shopping and Bing | | Freshness | Searches run at request time; `period=h`, `d`, `w`, `m` or `y` keeps recent pages only | | Ready-made tool code | [MCP server](https://serpcrypto.com/docs/mcp/), [OpenClaw](https://serpcrypto.com/docs/openclaw/), [LangChain](https://serpcrypto.com/docs/langchain/), [OpenAI function calling](https://serpcrypto.com/docs/openai/), [n8n](https://serpcrypto.com/docs/n8n/); tool name `google_search` | | Billing | 1 credit = 1 successful search request, which returns one page of results on any supported engine. A request that fails is not charged | | Price | $7.50 for 10,000 searches; $56.25 for 100,000 searches | | Keys | API keys you can create and revoke at any time from your dashboard | | Payment | BTC, USDT, USDC, ETH, SOL, LTC; no KYC, no card, no subscription | ## Should my agent use a SERP API or an AI search API? Use a SERP API when the agent must see what Google or Bing shows people; use an AI search API when you want the provider to pick and extract pages for you. Both are valid tools, and many agents use one of each. A **SERP API** such as SerpCrypto returns the engine's own results page: positions, links, snippets, ads, People Also Ask and the AI Overview. Your code or your model decides what to read next. **AI search APIs** such as Tavily, Exa, the Brave Search API and Perplexity's Search API select results with their own systems and often return extracted page text ready for a model. | | SERP API (SerpCrypto) | AI search API | |---|---|---| | Source of the ranking | Google or Bing, as people see it | The provider's own index or ranking | | Typical output | Results page: links, snippets, blocks | Selected pages, often with extracted text | | Local and country views | `gl`, `hl`, `location`, `device` | Varies by provider | | Best for | Citations people recognise, SEO-aware agents, local answers, news | Deep reading of a few pages with less code | We rank the options with dated prices on [best search APIs for AI agents](https://serpcrypto.com/best/search-apis-for-ai-agents/). ## When do Google results matter for an agent? They matter when the answer depends on what the open web shows today, in a given place. Four cases come up again and again: - **Citations people trust:** results that rank on Google are pages users already know, which makes an agent's sources easier to check. - **Freshness:** `top_stories` and `period=d` surface what was published today, which a static index may miss. - **Local answers:** opening hours, nearby services and local news change with `location`. - **What Google's AI cites:** Similarweb data reported by TechCrunch on 27 July 2026 put AI Overviews on 43% of searches. The `ai_overview.references` list shows which pages Google's own AI relied on, a useful second opinion for an agent. ## How do I give search results to a model without wasting tokens? Send the model only the fields it needs to reason and cite: title, link, snippet and date, for the first five to eight results. Drop ads, thumbnails, sitelinks and metadata. Number each result so the model can cite "[3]" and your code can map it back to a link. The function below does exactly that, and appends the AI Overview sources when Google showed an overview. Pass its output as the tool result. *Python: turn one search into a short, citable context for a model* ```python import os import requests API = "https://serpcrypto.com/api/v1/search" HEADERS = {"Authorization": "Bearer " + os.environ["SERPCRYPTO_API_KEY"]} def search_context(query, gl="us", hl="en", k=6, period=None): """Run one search and return numbered results as plain text for an LLM.""" params = {"engine": "google", "q": query, "gl": gl, "hl": hl} if period: params["period"] = period resp = requests.get(API, headers=HEADERS, params=params, timeout=60) resp.raise_for_status() data = resp.json() lines = [] for i, r in enumerate(data.get("organic_results", [])[:k], start=1): date = " (" + r["date"] + ")" if r.get("date") else "" lines.append("[%d] %s%s\n%s\n%s" % (i, r["title"], date, r.get("snippet", ""), r["link"])) overview = data.get("ai_overview") or {} refs = [ref["link"] for ref in overview.get("references", [])][:5] if refs: lines.append("Sources cited by Google's AI Overview: " + ", ".join(refs)) return "\n\n".join(lines) or "No results." if __name__ == "__main__": print(search_context("EU AI Act obligations for general-purpose models", period="m")) ``` ## Which agent frameworks can call SerpCrypto? Any framework that can make an HTTPS request can call the API, and the docs include copy-paste tool code for the most common ones. These are code samples, not published packages: you keep them in your own project. - **[MCP server](https://serpcrypto.com/docs/mcp/):** a small Python server that gives Claude Desktop, Cursor and other MCP clients a `google_search` tool. - **[OpenClaw](https://serpcrypto.com/docs/openclaw/):** the same MCP server added with `openclaw mcp add`, so the agent gets a `google_search` tool. - **[LangChain](https://serpcrypto.com/docs/langchain/):** a tool function an agent can call, plus a LlamaIndex variant. - **[OpenAI function calling](https://serpcrypto.com/docs/openai/):** the JSON schema of a `google_search` tool and the loop that runs it; the same schema works with other tool-calling models. - **[n8n](https://serpcrypto.com/docs/n8n/):** an HTTP Request node that calls the endpoint inside an AI agent workflow. The tool is called `google_search` because Claude Code and OpenClaw already have a built-in tool named `web_search`. OpenClaw's web tool documentation, read on 22 October 2026, lists built-in providers such as Brave, Perplexity, Exa, Tavily and SearXNG; none of them returns Google results pages. ## How do I make the agent cite its sources? Keep the number and the link of each result together, and tell the model to cite numbers. A short instruction is enough: "Answer from the numbered results only. After each claim, cite the result number in brackets. If the results do not answer the question, say so." After the answer comes back, replace each "[n]" with the link of result n. Your users then see real URLs, and you can check that every cited number exists. Results without a source should be treated as the model's own opinion. ## How many searches does an agent use, and what does it cost? It depends on how many questions the agent answers and how many searches it runs per question. A research agent that runs 3 to 5 searches per question and answers 200 questions a day uses 600 to 1,000 searches a day, or 18,000 to 30,000 a month. At that pace, 100,000 searches ($56.25) last three to five months. Because credits never expire, an agent that is only used in bursts keeps its budget for the next run. A separate API key per agent shows which one spends what. ## Why pay per search, in crypto, for an agent? Agents are often side projects, prototypes or tools that run in bursts, and a monthly plan that resets wastes money on them. Prepaid credits that never expire fit that pattern better. The official options are also closing. Microsoft retired the Bing Search APIs on 11 August 2025, and Google's Custom Search JSON API is closed to new customers, with existing customers given until 1 January 2027 to move. SerpCrypto keeps the results page available behind one endpoint, paid in BTC, USDT, USDC, ETH, SOL, LTC, with no ID, no selfie and no card. Crypto is not new to agent tools. On 22 October 2026, Linkup, You.com, Tavily and Exa documented per-request payments in USDC through the x402 protocol, without an account. Those APIs return results from their own systems, not Google or Bing pages; SerpCrypto covers that second need with a prepaid balance. ## Which pack suits an AI agent? A prototype fits in 10,000 searches ($7.50); an agent in daily use usually needs 25,000 searches to 100,000 searches, as of October 2026. Every engine costs one credit per search. | Plan | Quantity | Price | Price per unit | | --- | --- | --- | --- | | Test | 10,000 searches | $7.50 | $0.75/1K searches | | Starter | 25,000 searches | $18.75 | $0.75/1K searches | | Build | 100,000 searches | $56.25 | $0.56/1K searches | | Grow | 250,000 searches | $140.63 | $0.56/1K searches | In every plan: Every engine at 1 credit per search · Failed requests not charged · Credits never expire · Country, language, city and device targeting **Before you pay** - Minimum top-up $25. Top up once in crypto, then buy any plan from your balance. - Unused balance stays on your account and never expires, and it pays for search credits. When a purchase costs less than the minimum top-up, the part added only to reach the minimum is refundable on request to support within 30 days of the payment, in USDT or USDC, as long as it has not been spent. The rest of the balance is not paid back, except where the law requires it. - Credited once your payment is checked on the blockchain. Typical network confirmation time: BTC ≈ 20 min · USDT TRC-20 ≈ 1 min · USDT ERC-20 ≈ 5 min · USDC ERC-20 ≈ 5 min · ETH ≈ 5 min · SOL ≈ 1 min · LTC ≈ 15 min ## Search API for AI agents: common questions ### Does SerpCrypto return the full text of each page? No. It returns the results page: titles, links, snippets and the other blocks Google or Bing showed. If your agent needs the text of a page, fetch the link yourself and respect that site's terms. ### Can I use SerpCrypto with Claude, ChatGPT or Gemini models? Yes. Any model that supports tool or function calling can use it through your code. The docs have ready-made code for [MCP](https://serpcrypto.com/docs/mcp/), [LangChain](https://serpcrypto.com/docs/langchain/) and [OpenAI-style function calling](https://serpcrypto.com/docs/openai/). ### Is there an official SerpCrypto SDK or MCP package? No. The integrations are short code samples you copy into your project. The API itself is one HTTPS endpoint, so a published package is not needed. ### How fresh are the results? Searches run at request time, so the agent sees the results page at that moment. Add `period=d` or `period=h` to keep only pages or stories from the past day or hour. ### Can an agent get its own key and budget? Yes. API keys you can create and revoke at any time from your dashboard. Create one key per agent; usage per key shows which agent spends your credits. ### Do I need an ID or a credit card? No. SerpCrypto asks for no ID, no selfie, no company documents and no card. You top up from $25 in BTC, USDT, USDC, ETH, SOL, LTC. ## Sources 1. [Google's AI search is rapidly becoming the default, new data shows](https://techcrunch.com/2026/07/27/googles-ai-search-is-rapidly-becoming-the-default-new-data-shows/) — TechCrunch (2026-07-27) 2. [Web tools (web search providers)](https://docs.openclaw.ai/tools/web) — OpenClaw docs (checked 2026-10-22) 3. [Bing Search API retirement](https://learn.microsoft.com/en-us/lifecycle/announcements/bing-search-api-retirement) — Microsoft Learn (checked 2026-10-22) 4. [Custom Search JSON API overview](https://developers.google.com/custom-search/v1/overview) — Google for Developers (checked 2026-10-22) 5. [x402 payments](https://docs.linkup.so/pages/documentation/platform/x402) — Linkup docs (checked 2026-10-22) 6. [Machine payments](https://you.com/docs/administration/machine-payments) — You.com docs (checked 2026-10-22) 7. [Tavily on x402](https://tavily.com/blog/tavily-on-x402) — Tavily (checked 2026-10-22) 8. [x402 guide](https://exa.ai/docs/reference/x402-guide) — Exa docs (checked 2026-10-22) --- SerpCrypto: SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. Prepaid search credits from $0.30/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto. --- # Brand monitoring with a SERP API: news, web results and lookalike pages > Brand monitoring API: new Google News stories and web results about your names, hourly or daily, as JSON. From $0.30/1K searches, no KYC, pay in crypto. - URL: https://serpcrypto.com/use-cases/brand-monitoring/ - Last updated: 2026-10-22 - Publisher: SerpCrypto (https://serpcrypto.com) Know when a story names your company, what ranks when people search your brand, and which unknown sites appear next to your name. One API, a few queries, a schedule. **Quick answer — How can I monitor news and search results about my brand with an API?** Schedule a few searches: Google News with `period=h` for new stories, and Google web search for what ranks on your brand name. **SerpCrypto** returns both as JSON for one credit per search, from $0.30/1K searches as of October 2026. Twenty names checked hourly in Google News is 14,400 searches a month. No KYC; pay in crypto. ## Brand monitoring with SerpCrypto | Attribute | Value | | --- | --- | | New stories | `engine=google_news`, `period=h` or `d`: headline, source, date, link | | What ranks on your name | `engine=google`: organic results, top stories, People Also Ask, knowledge panel | | Copies of your images | `engine=google_images`: pages that host similar pictures | | Markets | `gl` and `hl` per country edition | | Cost | 1 credit per search; 25,000 searches for $18.75 | | Payment | BTC, USDT, USDC, ETH, SOL, LTC; no KYC, no card | ## Which searches should a brand monitor run? Three kinds of searches cover most needs, each on its own schedule. 1. **News mentions, often:** your company, product and executive names in quotes on Google News, with `period=h` for hourly checks or `period=d` for a daily digest. 2. **Your brand page, daily:** the plain brand name on Google web search. Watch who holds the top ten, the knowledge panel and the People Also Ask questions. 3. **Risky combinations, weekly:** your brand with words such as "login", "support", "refund" or "download", to spot pages that imitate you. Keep each query fixed, so changes in the results mean something. *Python: daily check of who ranks on your brand name* ```python import json import os from pathlib import Path from urllib.parse import urlparse import requests API = "https://serpcrypto.com/api/v1/search" HEADERS = {"Authorization": "Bearer " + os.environ["SERPCRYPTO_API_KEY"]} BRAND = "acme robotics" KNOWN = {"acme.example", "wikipedia.org", "linkedin.com"} STATE = Path("brand-serp.json") params = {"engine": "google", "q": BRAND, "gl": "us", "hl": "en"} resp = requests.get(API, headers=HEADERS, params=params, timeout=60) resp.raise_for_status() hosts = [urlparse(r["link"]).netloc.lower().removeprefix("www.") for r in resp.json().get("organic_results", [])] before = json.loads(STATE.read_text()) if STATE.exists() else [] for host in hosts: if host not in before: flag = "" if any(host.endswith(k) for k in KNOWN) else " <- unknown site" print("new in top 10:", host + flag) STATE.write_text(json.dumps(hosts)) ``` ## How do I spot sites that imitate my brand? Look for unknown domains that rank on your brand name combined with words people use when they need help or want to log in. A new site in the top results for "yourbrand support" or "yourbrand login" deserves a manual check. The API only shows you that a page ranks; it does not judge it. Review each new domain by hand, and if a site impersonates you, report it to the search engine, the host and the registrar. Our [Acceptable Use Policy](https://serpcrypto.com/acceptable-use/) welcomes brand protection work. ## What does the results page tell me about my reputation? The first page for your brand name is what most people see before they decide to trust you. Four parts matter: - **Top results:** your own pages, reviews sites, press and forums, in the order Google shows them. - **Top stories:** news cards that appear when your name is in the news. - **People Also Ask:** the questions people ask about you, which often reveal doubts you should answer on your own site. - **AI Overview:** when Google shows one for your brand, its `references` list shows which pages its summary relied on. Store these blocks daily; a change in any of them is worth a look. ## How many searches does brand monitoring use? Count queries × editions × checks. Twenty names checked hourly on Google News is 20 × 24 × 30 = 14,400 searches a month. Your brand page checked daily in 3 countries adds 90, and 10 risky combinations weekly add 40. That whole setup is about 14,530 searches a month, which 25,000 searches ($18.75) cover for more than six weeks. A daily news digest instead of hourly checks divides the news part by 24. ## Which pack suits brand monitoring? A daily setup fits in 10,000 searches ($7.50); hourly news checks for several names need 25,000 searches or more, as of October 2026. Credits never expire. | Plan | Quantity | Price | Price per unit | | --- | --- | --- | --- | | Test | 10,000 searches | $7.50 | $0.75/1K searches | | Starter | 25,000 searches | $18.75 | $0.75/1K searches | | Build | 100,000 searches | $56.25 | $0.56/1K searches | In every plan: Every engine at 1 credit per search · Failed requests not charged · Credits never expire · Country, language, city and device targeting **Before you pay** - Minimum top-up $25. Top up once in crypto, then buy any plan from your balance. - Unused balance stays on your account and never expires, and it pays for search credits. When a purchase costs less than the minimum top-up, the part added only to reach the minimum is refundable on request to support within 30 days of the payment, in USDT or USDC, as long as it has not been spent. The rest of the balance is not paid back, except where the law requires it. - Credited once your payment is checked on the blockchain. Typical network confirmation time: BTC ≈ 20 min · USDT TRC-20 ≈ 1 min · USDT ERC-20 ≈ 5 min · USDC ERC-20 ≈ 5 min · ETH ≈ 5 min · SOL ≈ 1 min · LTC ≈ 15 min ## Brand monitoring API: common questions ### Can I monitor social networks with this API? Only what search engines show about them. The API returns Google and Bing results, which can include social pages that rank. It does not read social network feeds directly. ### How quickly will I see a new story? At your first check after Google News lists it. With hourly checks and `period=h`, a newly listed story shows up at the next hourly run, so the delay is mostly Google's own listing time plus up to one hour. ### Can I monitor several countries? Yes. Run each query with the `gl` and `hl` of each country edition. Each edition is one search, so three countries triple the news part of the budget. ### Does the API tell me if a story is positive or negative? No. It returns headlines, snippets and links. Classify them yourself, by hand or with a language model; the [AI agents page](https://serpcrypto.com/use-cases/ai-agents/) shows how to pass results to a model. ### How do I pay? Top up from $25 in BTC, USDT, USDC, ETH, SOL, LTC and buy a pack from your balance. SerpCrypto asks for no ID, no selfie and no card. --- SerpCrypto: SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. Prepaid search credits from $0.30/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto. --- # Keyword research with a SERP API: questions, related searches and intent > Keyword research API: expand seed keywords with People Also Ask and related searches, and read the SERP features of each query. From $0.30/1K searches, no KYC. - URL: https://serpcrypto.com/use-cases/keyword-research/ - Last updated: 2026-10-22 - Publisher: SerpCrypto (https://serpcrypto.com) The results page tells you what people ask next, what Google thinks a query means and who already ranks. Read it as JSON and grow a seed list into a topic map. **Quick answer — How can I use a SERP API for keyword research?** Search each seed keyword once and collect two lists from the JSON: `people_also_ask` questions and `related_searches` queries. Search those in turn for a second round. **SerpCrypto** returns both lists, plus the SERP features of each query, for one credit per search, from $0.30/1K searches as of October 2026. 200 seeds expanded twice is about 2,000 searches. No KYC; pay in crypto. ## Keyword research with SerpCrypto | Attribute | Value | | --- | --- | | Question ideas | `people_also_ask`: question, answer snippet, source link | | Query ideas | `related_searches`: the suggestions at the bottom of the page | | Intent signals | Which blocks appear: AI Overview, local pack, shopping, top stories, videos | | Competition | The domains in `organic_results` for each query | | Not provided | Search volume, keyword difficulty scores, autocomplete suggestions | | Cost | 1 credit per search; 10,000 searches for $7.50 | ## What can the results page tell me about a keyword? It tells you what people want when they type it, as Google reads it. The blocks on the page are the clearest clue: | Block on the page | What it suggests | |---|---| | `local_results` (map pack) | Local intent: people want a place nearby | | `shopping_results` | Buying intent for a product | | `top_stories` | News intent: freshness matters | | `inline_videos` | A video answer is expected | | `ai_overview` | An informational query Google answers itself | | `people_also_ask` | Related questions worth answering on the same page | The domains in `organic_results` show who you compete with, and whether the top results are big brands, forums or small sites like yours. ## How do I expand a seed list without wasting searches? Expand in rounds and stop early. Round one searches every seed; round two searches only the new questions and queries that look relevant; a third round is rarely worth it. - **Deduplicate** new terms before searching them, ignoring case and punctuation. - **Filter** by words that must appear, such as your product category, to stay on topic. - **Cap** the number of searches per seed, so one broad seed cannot eat the budget. The code below applies all three rules with a fixed search budget. *Python: two-round expansion with a search budget* ```python import os import requests API = "https://serpcrypto.com/api/v1/search" HEADERS = {"Authorization": "Bearer " + os.environ["SERPCRYPTO_API_KEY"]} def ideas(query, gl="us", hl="en"): params = {"engine": "google", "q": query, "gl": gl, "hl": hl} resp = requests.get(API, headers=HEADERS, params=params, timeout=60) resp.raise_for_status() data = resp.json() found = [q["question"] for q in data.get("people_also_ask", [])] found += [r["query"] for r in data.get("related_searches", [])] return found def expand(seeds, must_contain, budget=100, rounds=2): seen = {s.lower() for s in seeds} frontier, used = list(seeds), 0 for _ in range(rounds): nxt = [] for term in frontier: if used >= budget: return sorted(seen) used += 1 for idea in ideas(term): key = idea.lower().strip(" ?") if key not in seen and any(w in key for w in must_contain): seen.add(key) nxt.append(idea) frontier = nxt return sorted(seen) if __name__ == "__main__": for kw in expand(["standing desk"], must_contain=["desk"], budget=40): print(kw) ``` ## Does SerpCrypto give search volume or keyword difficulty? No. SerpCrypto returns results pages, not search volumes or difficulty scores. Volumes come from tools that measure demand, such as Google's own Keyword Planner inside Google Ads, or third-party SEO suites. The two kinds of data work well together: a volume tool tells you how many people search a term, and the results page tells you what they expect to find and who already ranks. Autocomplete suggestions are not part of our engines either. ## How do I turn questions into content? Group the questions you collect by the page that should answer them. People Also Ask questions that share a topic belong on one page, each as a heading with a short, direct answer under it. Then search each draft page's main query again after publication. If your page starts to appear in `people_also_ask` sources, in the AI Overview references or in the top results, the grouping worked. Our [rank tracking guide](https://serpcrypto.com/use-cases/rank-tracking/) shows how to follow that over time. ## How many searches does a keyword research project use? Count one search per term you look up. A project with 200 seeds and a second round on the 1,800 most relevant new ideas uses about 2,000 searches. Checking each idea in 3 countries triples that. Research is usually a one-off or a quarterly job, so prepaid credits that never expire fit it well. 10,000 searches ($7.50) cover several projects of that size, and the balance waits for the next one. ## Which pack suits keyword research? One project usually fits in 10,000 searches ($7.50); agencies running research for many clients often start at 100,000 searches ($56.25), as of October 2026. | Plan | Quantity | Price | Price per unit | | --- | --- | --- | --- | | Test | 10,000 searches | $7.50 | $0.75/1K searches | | Starter | 25,000 searches | $18.75 | $0.75/1K searches | | Build | 100,000 searches | $56.25 | $0.56/1K searches | In every plan: Every engine at 1 credit per search · Failed requests not charged · Credits never expire · Country, language, city and device targeting **Before you pay** - Minimum top-up $25. Top up once in crypto, then buy any plan from your balance. - Unused balance stays on your account and never expires, and it pays for search credits. When a purchase costs less than the minimum top-up, the part added only to reach the minimum is refundable on request to support within 30 days of the payment, in USDT or USDC, as long as it has not been spent. The rest of the balance is not paid back, except where the law requires it. - Credited once your payment is checked on the blockchain. Typical network confirmation time: BTC ≈ 20 min · USDT TRC-20 ≈ 1 min · USDT ERC-20 ≈ 5 min · USDC ERC-20 ≈ 5 min · ETH ≈ 5 min · SOL ≈ 1 min · LTC ≈ 15 min ## Keyword research API: common questions ### How many People Also Ask questions does one search return? The questions Google shows on the page when it loads, usually a handful. Google reveals more when a person clicks a question; to go further, search the new questions themselves in a second round. ### Can I research keywords in other languages? Yes. Set `gl` and `hl` for the market, for example `gl=fr&hl=fr` for France. Questions and related searches come back in that language. ### Is this the same as Google autocomplete? No. Autocomplete suggestions are not part of the SerpCrypto engines. Related searches and People Also Ask come from the results page itself, after the search runs. ### Can I see which SERP features a keyword triggers in bulk? Yes. Search each keyword once and record which keys exist in the JSON, such as `ai_overview`, `local_results` or `shopping_results`. One search per keyword is enough. ### How do I pay? Top up from $25 in BTC, USDT, USDC, ETH, SOL, LTC and buy a pack from your balance. No ID, no selfie and no card. --- SerpCrypto: SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. Prepaid search credits from $0.30/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto. --- # Local SEO API: see the results and the map pack from any city > Local SEO API for 2026: check Google results and the map pack from any city with one parameter, from $0.30/1K searches. JSON output, no KYC, crypto. - URL: https://serpcrypto.com/use-cases/local-seo/ - Last updated: 2026-10-22 - Publisher: SerpCrypto (https://serpcrypto.com) Local rankings change from one city to the next, and the map pack changes even more. Set the location parameter and read what a searcher there sees, without travelling or guessing. **Quick answer — How can I check local Google results and the map pack for another city?** Send the query to a SERP API with a city-level `location`, such as `Denver, Colorado, United States`. **SerpCrypto** returns the Google results page that location gets, with the map pack in `local_results` (name, address, rating, review count, category), for one credit per search from $0.30/1K searches as of October 2026. No KYC; pay in crypto. ## Local SEO with SerpCrypto | Attribute | Value | | --- | --- | | Location parameter | `location=City, Region, Country`, plus `gl` and `hl` | | Map pack data | `local_results`: position, title, address, rating, reviews, type | | Full Maps listings | `engine=google_maps`: also phone, website, opening hours, coordinates | | Devices | Desktop or mobile (`device=mobile`) | | Price | 1 credit per search; 25,000 searches for $18.75, 100,000 searches for $56.25 | | Payment | BTC, USDT, USDC, ETH, SOL, LTC; no KYC | ## Why do local results differ from one city to the next? For queries with local intent, Google ranks businesses by relevance, distance and popularity, so the searcher's location changes the answer. A query such as "emergency plumber" or "pediatric dentist" shows different map packs in two suburbs of the same metro area, and often different organic results too. This is why a ranking checked from your office, or from a generic national search, tells you little about a client's market. You need the page as it looks from each place that matters to the business. ## What is the difference between the map pack and Google Maps results? The map pack is the block of about three businesses with a small map inside the normal Google results page. Google Maps is the separate map search, which lists many more places with more detail. | | Map pack (`engine=google`) | Maps search (`engine=google_maps`) | |---|---|---| | Where it appears | Inside the web results page | On the Maps search | | Businesses per search | Usually 3 | A full list | | Fields | Name, address, rating, reviews, category | Same, plus phone, website, hours, coordinates | | Best for | Visibility in web search | Audits, competitor lists, data on places | Both cost one credit per search. The [Google Maps API page](https://serpcrypto.com/google-maps-api/) covers the Maps engine in detail. ## How do I track a business across several cities? Pick the queries and the places, then run one search per query and place. Record two numbers for each: the map pack position of the business, and its organic position. 1. **List the places** where the business serves clients: city names for a regional firm, neighbourhoods or suburbs for a single shop. 2. **Write location strings** in the `City, Region, Country` form. The [locations page](https://serpcrypto.com/docs/locations/) shows the format. 3. **Choose a rhythm:** weekly is enough for most local businesses; daily during a launch or after a move. 4. **Store the raw JSON** for each check, so you can answer new questions later without paying again for old searches. *Python: map pack position of a business in five cities* ```python import os import requests API = "https://serpcrypto.com/api/v1/search" HEADERS = {"Authorization": "Bearer " + os.environ["SERPCRYPTO_API_KEY"]} CITIES = [ "Denver, Colorado, United States", "Aurora, Colorado, United States", "Boulder, Colorado, United States", "Lakewood, Colorado, United States", "Littleton, Colorado, United States", ] def map_pack_position(business, query, location): params = {"engine": "google", "q": query, "gl": "us", "hl": "en", "location": location} resp = requests.get(API, headers=HEADERS, params=params, timeout=60) resp.raise_for_status() for place in resp.json().get("local_results", []): if business.lower() in place.get("title", "").lower(): return place.get("position") return None for city in CITIES: print(city, "->", map_pack_position("Example Plumbing Co", "emergency plumber", city)) ``` ## How much does local rank tracking cost? Count one search per query, place and check. Ten queries in five cities checked weekly is 10 × 5 × 4 = 200 searches a month; the same set checked daily is 1,500. Adding mobile doubles either figure. At that scale, 10,000 searches ($7.50) covers a single business for months. An agency with 40 clients, 15 queries and 4 places each, checked weekly, needs about 9,600 searches a month, which 100,000 searches ($56.25) covers for about ten months. Credits never expire, so seasonal clients cost nothing between campaigns. ## What else can a local SEO audit pull from the same data? The same JSON answers several audit questions without extra tools: - **Competitor list:** the businesses that hold the map pack in each place, with ratings and review counts. - **Category gaps:** the `type` that Google shows for the leaders compared with your client's category. - **Review gap:** how many reviews separate your client from the third business in the pack. - **Consistency of name and address:** whether the listing shows the address you expect in each search. - **Local questions:** People Also Ask questions that mention the city, useful for location pages. ## Which pack suits local SEO? A single business fits in 10,000 searches; an agency usually starts at 100,000 searches. All engines, including Google Maps, cost one credit per search. | Plan | Quantity | Price | Price per unit | | --- | --- | --- | --- | | Test | 10,000 searches | $7.50 | $0.75/1K searches | | Starter | 25,000 searches | $18.75 | $0.75/1K searches | | Build | 100,000 searches | $56.25 | $0.56/1K searches | In every plan: Every engine at 1 credit per search · Failed requests not charged · Credits never expire · Country, language, city and device targeting **Before you pay** - Minimum top-up $25. Top up once in crypto, then buy any plan from your balance. - Unused balance stays on your account and never expires, and it pays for search credits. When a purchase costs less than the minimum top-up, the part added only to reach the minimum is refundable on request to support within 30 days of the payment, in USDT or USDC, as long as it has not been spent. The rest of the balance is not paid back, except where the law requires it. - Credited once your payment is checked on the blockchain. Typical network confirmation time: BTC ≈ 20 min · USDT TRC-20 ≈ 1 min · USDT ERC-20 ≈ 5 min · USDC ERC-20 ≈ 5 min · ETH ≈ 5 min · SOL ≈ 1 min · LTC ≈ 15 min ## Local SEO API: common questions ### Can I check results for a ZIP code or a neighbourhood? Use the most precise place name that Google recognises, such as a suburb or a district written as `District, City, Region, Country`. City-level names are the most reliable; very small areas may resolve to the nearest larger place. ### Is the map pack the same on mobile and desktop? Not always. The phone layout can show the pack higher on the page and with different businesses. Run a second search with `device=mobile` when mobile matters to the client; it is a separate credit. ### Can I get phone numbers and opening hours? Yes, from the Maps engine. Set `engine=google_maps` and each place in `local_results` includes phone, website, hours and coordinates when Google shows them. See the [Google Maps API](https://serpcrypto.com/google-maps-api/). ### Does SerpCrypto need my clients' Google Business Profile access? No. The API reads public search results only. It does not connect to Business Profile accounts, so you never share client logins with us. ### How do I pay without a company card? Top up from $25 in BTC, USDT, USDC, ETH, SOL, LTC, then buy a pack from your balance. SerpCrypto asks for no ID, no selfie and no company documents. ## Sources 1. [Tips to improve your local ranking on Google](https://support.google.com/business/answer/7091) — Google Business Profile Help (checked 2026-10-22) --- SerpCrypto: SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. Prepaid search credits from $0.30/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto. --- # Price monitoring with a SERP API: offers, merchants and prices by country > Price monitoring API: track Google Shopping offers, merchants and prices by country as JSON, with alert code and budgets, from $0.30/1K searches. No KYC. - URL: https://serpcrypto.com/use-cases/price-monitoring/ - Last updated: 2026-10-22 - Publisher: SerpCrypto (https://serpcrypto.com) Watch what shoppers see when they search for your products: who sells them, at what price, with what delivery. Store a daily snapshot and get an alert when a competitor moves. **Quick answer — How can I monitor competitor prices on Google Shopping with an API?** Run one Google Shopping search per product and country each day, store the offers, and compare them with yesterday's. **SerpCrypto** returns each offer as JSON with merchant, price as a number, currency and delivery line, for one credit per search, from $0.30/1K searches as of October 2026. 500 products in 3 countries checked daily is 45,000 searches a month. No KYC; pay in crypto. ## Price monitoring with SerpCrypto | Attribute | Value | | --- | --- | | Engine | `google_shopping` (offers); `google` for retailer pages in organic results | | Price fields | `price` (as shown), `extracted_price` (number), `currency`, `delivery` | | Seller field | `source` (merchant name) | | Country | `gl` sets the market and the currency; `hl` the language | | Cost | 1 credit per product, country and check; failed searches free | | Price of credits | 100,000 searches for $56.25; 500,000 searches for $281.25 | ## What should a price monitor record each day? Record one row per offer and day: date, country, product query, merchant, numeric price, currency, delivery line and position. With those eight columns you can answer most pricing questions later without paying for the same search twice. - **Minimum price per country:** the smallest `extracted_price` for a product and day. - **Your rank among sellers:** the `position` of your own `source` name. - **Price changes:** the difference with the previous day for each merchant. - **New sellers:** merchants that appear for the first time. - **Delivery promises:** free shipping or delivery times that win the click without a lower price. ## How do I match search results to my own products? Search with the words a shopper would use, usually brand plus model, and keep the query fixed from day to day. Then match the offer titles you get back to your catalogue once, by hand or with simple rules, and store the mapping. Avoid queries that are too broad, such as a category name, because the offers change with every search. Avoid queries that are too narrow, such as an internal SKU, because shoppers never type them and Google may return nothing useful. *Python: daily snapshot and an alert when a competitor undercuts you* ```python import csv import datetime import os import requests API = "https://serpcrypto.com/api/v1/search" HEADERS = {"Authorization": "Bearer " + os.environ["SERPCRYPTO_API_KEY"]} OUR_STORE = "Example Store" PRODUCTS = {"garmin forerunner 265": 349.99, "jabra elite 10": 199.00} today = datetime.date.today().isoformat() with open("prices.csv", "a", newline="", encoding="utf-8") as fh: writer = csv.writer(fh) for query, our_price in PRODUCTS.items(): params = {"engine": "google_shopping", "q": query, "gl": "us", "hl": "en"} resp = requests.get(API, headers=HEADERS, params=params, timeout=60) resp.raise_for_status() for offer in resp.json().get("shopping_results", []): price = offer.get("extracted_price") seller = offer.get("source", "") writer.writerow([today, "us", query, seller, price, offer.get("currency", ""), offer.get("position")]) if price and seller != OUR_STORE and price < our_price: print("ALERT:", seller, "sells", query, "at", price, "below", our_price) ``` ## How often should I check prices? Daily is enough for most catalogues; twice a day suits sales seasons and fast categories such as electronics. Hourly checks rarely pay off on Google Shopping, because merchants' feeds do not change that often. A mixed rhythm keeps costs down: check your 50 best sellers twice a day and the rest of the catalogue once a week. Because SerpCrypto credits never expire, a quiet month does not waste a pack, and you can buy ahead of a sales season. ## How many searches does price monitoring use? Searches per month for common setups, counted as 30 daily checks or 4 weekly checks. | Setup | Searches per month | Pack that covers a month | | --- | --- | --- | | 100 products, 1 country, daily | 3,000 | 10,000 searches ($7.50) last about 3 months | | 500 products, 3 countries, daily | 45,000 | 100,000 searches ($56.25) last about 2 months | | 50 best sellers twice a day + 2,000 products weekly | 11,000 | 25,000 searches ($18.75) last about 2 months | | 5,000 products, 2 countries, daily | 300,000 | 500,000 searches ($281.25) last about 7 weeks | ## Can I watch retailer pages in normal search results too? Yes. Some shoppers never open the Shopping tab. Run the same product query with `engine=google` and read `organic_results` and `shopping_results`: you see which retailers rank in the main results and whether Google shows a product block there. That second search costs one more credit. Many teams run it weekly rather than daily, since organic rankings move more slowly than prices. ## Which pack suits price monitoring? Small catalogues fit in 25,000 searches; a few hundred products in several countries usually need 100,000 searches or more, as of October 2026. Credits never expire. | Plan | Quantity | Price | Price per unit | | --- | --- | --- | --- | | Starter | 25,000 searches | $18.75 | $0.75/1K searches | | Build | 100,000 searches | $56.25 | $0.56/1K searches | | Grow | 250,000 searches | $140.63 | $0.56/1K searches | | Scale | 500,000 searches | $281.25 | $0.56/1K searches | In every plan: Every engine at 1 credit per search · Failed requests not charged · Credits never expire · Country, language, city and device targeting **Before you pay** - Minimum top-up $25. Top up once in crypto, then buy any plan from your balance. - Unused balance stays on your account and never expires, and it pays for search credits. When a purchase costs less than the minimum top-up, the part added only to reach the minimum is refundable on request to support within 30 days of the payment, in USDT or USDC, as long as it has not been spent. The rest of the balance is not paid back, except where the law requires it. - Credited once your payment is checked on the blockchain. Typical network confirmation time: BTC ≈ 20 min · USDT TRC-20 ≈ 1 min · USDT ERC-20 ≈ 5 min · USDC ERC-20 ≈ 5 min · ETH ≈ 5 min · SOL ≈ 1 min · LTC ≈ 15 min ## Price monitoring API: common questions ### Is the price including tax and shipping? The API returns the price exactly as Google Shopping shows it in that country, plus the `delivery` line when Google shows one. Tax display rules differ by country, so compare within one market first. ### Can I monitor prices in several currencies? Yes. Each country edition returns its local currency in the `currency` field. Convert with your own exchange rates if you need one currency across markets. ### Can I get the price from the merchant's own page? The API returns the price Google shows and a link to the offer. Reading the merchant's page is a separate step that you do yourself, under that site's terms. ### What happens if a search fails during the nightly run? You get an error code and the failed search is not charged. Retry it later; the [errors page](https://serpcrypto.com/docs/errors/) shows a retry loop with backoff. ### How do I pay? Top up from $25 in BTC, USDT, USDC, ETH, SOL, LTC and buy a pack from your balance. No ID, no selfie and no card are needed. --- SerpCrypto: SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. Prepaid search credits from $0.30/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto. --- # Rank tracking API: how many searches you need, and what they cost > Rank tracking API maths for 2026: searches per keyword, top 10 vs top 100 after num=100, and the cost from $0.30/1K searches. No KYC, pay in crypto. - URL: https://serpcrypto.com/use-cases/rank-tracking/ - Last updated: 2026-10-22 - Publisher: SerpCrypto (https://serpcrypto.com) Since Google dropped num=100, the cost of rank tracking depends on how deep you look. This page gives the formula, ready-made budgets and code that stops paging as soon as it finds your domain. **Quick answer — How many API searches does rank tracking need, and what does it cost?** Searches per month = keywords × pages checked × checks per month × locations × devices. A daily top-10 check of 1,000 keywords is 30,000 searches a month; a daily top-100 check is 300,000, because Google now serves about 10 results per page. With **SerpCrypto**, a no-KYC SERP API paid in crypto, 100,000 searches cost $56.25 and credits never expire (October 2026). ## Rank tracking with SerpCrypto: the numbers | Attribute | Value | | --- | --- | | Cost of one check | 1 credit per results page; About 10 organic results per page, as Google has served since it stopped honouring num=100 in September 2025; each extra page is one more credit | | Price per 1,000 searches | $0.75/1K searches at 10,000 searches, $0.56/1K searches at 100,000 searches, $0.38/1K searches at 1,000,000 searches | | Failed searches | Not charged | | Targeting | Country (gl), language (hl), Google domain and a city-level location string, on desktop or mobile | | Extra signals in the same credit | AI Overview citations, People Also Ask, local pack, ads, top stories | | Credit expiry | Prepaid search credits never expire: unused searches stay on your account until you use them | ## How do I calculate the searches my rank tracker needs? Multiply five numbers: keywords, results pages per keyword, checks per month, locations and devices. One results page holds about 10 organic results, so tracking the top 30 means 3 pages per keyword per check. `searches per month = keywords × pages × checks per month × locations × devices` Two levers cut the bill without losing much signal. First, stop paging as soon as you find the domain you track: a page ranked 4th never needs pages 2 to 10. Second, check long-tail keywords weekly and money keywords daily. The code further down does the first automatically. ## What does a typical rank tracking setup cost? Monthly searches for common setups, and the smallest SerpCrypto pack that covers one month. A month is counted as 30 daily checks or 4 weekly checks. | Setup | Searches per month | Pack that covers a month | | --- | --- | --- | | 100 keywords, top 10, daily, 1 location | 3,000 | 10,000 searches ($7.50) lasts about 3 months | | 1,000 keywords, top 10, daily | 30,000 | 100,000 searches ($56.25) lasts about 3 months | | 1,000 keywords, top 30, weekly | 12,000 | 25,000 searches ($18.75) lasts about 2 months | | 1,000 keywords, top 100, weekly | 40,000 | 100,000 searches ($56.25) lasts about 2.5 months | | 5,000 keywords, top 10, daily | 150,000 | 250,000 searches ($140.63) lasts about 7 weeks | | 1,000 keywords, top 10, daily, 3 cities × desktop and mobile | 180,000 | 250,000 searches ($140.63) lasts about 6 weeks | | 10,000 keywords, top 100, weekly | 400,000 | 500,000 searches ($281.25) | Worst case: every keyword is checked to the last page. Stopping when the domain is found usually needs far fewer searches. ## How do I find my domain's position in the JSON? Walk through `organic_results` page by page and compare the host of each `link` with your domain. The `position` field is already the rank on the whole results list, so page 2 starts at about 11. Stop as soon as you find the domain, and record "not in the top N" when the last page you allow yields nothing. *Python: rank of a domain, stopping at the first match* ```python import os from urllib.parse import urlparse import requests API = "https://serpcrypto.com/api/v1/search" HEADERS = {"Authorization": "Bearer " + os.environ["SERPCRYPTO_API_KEY"]} def rank_of(domain, query, gl="us", hl="en", location=None, max_pages=3): """Return (position, url, searches_used); position is None if not found.""" params = {"engine": "google", "q": query, "gl": gl, "hl": hl} if location: params["location"] = location for page in range(1, max_pages + 1): params["page"] = page resp = requests.get(API, headers=HEADERS, params=params, timeout=60) resp.raise_for_status() for r in resp.json().get("organic_results", []): host = urlparse(r["link"]).netloc.lower() if host == domain or host.endswith("." + domain): return r["position"], r["link"], page return None, None, max_pages if __name__ == "__main__": pos, url, used = rank_of("example.com", "project management software", max_pages=3) print(pos, url, "searches used:", used) ``` ## Should I track AI Overview citations as well as positions? Yes, if your queries trigger AI Overviews, because a page can rank 7th and still be cited in the overview above position 1. The same search that gives you positions also returns `ai_overview.references` when Google shows an overview, at no extra credit. Store three things per keyword and day: your organic position, whether your domain is among the overview references, and which other domains are. Over a few weeks this shows which competitors Google's AI cites for your topics, which a position report alone cannot. ## How do I track rankings in several cities? Run the same query once per city with the `location` parameter, for example `Denver, Colorado, United States` and `Miami, Florida, United States`. Local intent queries, such as services, shops and clinics, can differ a lot from one city to the next, and the map pack in `local_results` changes even more. Each city is a separate search, so multiply your budget by the number of cities. The [local SEO page](https://serpcrypto.com/use-cases/local-seo/) covers map pack tracking, and [locations and languages](https://serpcrypto.com/docs/locations/) explains the location strings. ## Daily or weekly checks: which one do I need? Daily checks suit a few hundred keywords that drive revenue, or the weeks after a site migration or a Google update. Weekly checks suit long-tail sets, competitor monitoring and reporting to clients. A mixed schedule keeps the signal and divides the cost: 200 money keywords daily plus 2,000 long-tail keywords weekly, top 10 only, is 6,000 + 8,000 = 14,000 searches a month. Because SerpCrypto credits never expire, a quiet month does not waste what you bought. ## Which pack fits a rank tracker? Most rank trackers between 1,000 and 5,000 keywords fit between 100,000 searches and 500,000 searches. Every engine and every page costs one credit; credits never expire. | Plan | Quantity | Price | Price per unit | | --- | --- | --- | --- | | Starter | 25,000 searches | $18.75 | $0.75/1K searches | | Build | 100,000 searches | $56.25 | $0.56/1K searches | | Grow | 250,000 searches | $140.63 | $0.56/1K searches | | Scale | 500,000 searches | $281.25 | $0.56/1K searches | In every plan: Every engine at 1 credit per search · Failed requests not charged · Credits never expire · Country, language, city and device targeting **Before you pay** - Minimum top-up $25. Top up once in crypto, then buy any plan from your balance. - Unused balance stays on your account and never expires, and it pays for search credits. When a purchase costs less than the minimum top-up, the part added only to reach the minimum is refundable on request to support within 30 days of the payment, in USDT or USDC, as long as it has not been spent. The rest of the balance is not paid back, except where the law requires it. - Credited once your payment is checked on the blockchain. Typical network confirmation time: BTC ≈ 20 min · USDT TRC-20 ≈ 1 min · USDT ERC-20 ≈ 5 min · USDC ERC-20 ≈ 5 min · ETH ≈ 5 min · SOL ≈ 1 min · LTC ≈ 15 min ## Rank tracking API: common questions ### Why did my rank tracking costs go up in 2025? Google stopped honouring `num=100` around 10 to 14 September 2025. Before that, one request could return 100 results; now each page of about 10 results is a separate request. Tracking the top 100 therefore takes 10 times more searches than before. ### Can I track mobile and desktop rankings separately? Yes. Add `device=mobile` to get the phone results page. Mobile and desktop are separate searches, so tracking both doubles the searches for those keywords. ### Does SerpCrypto provide a rank tracking dashboard? No. SerpCrypto is an API: it returns the results page as JSON, and you store and chart the positions in your own tool, spreadsheet or database. The Python code on this page is a starting point. ### What happens if a search fails during my nightly job? You get an HTTP error with a code such as `search_failed` or `search_timeout`, and the failed request is not charged. Retry it later with exponential backoff; the [errors page](https://serpcrypto.com/docs/errors/) has a ready-made loop. ### Do I need an ID or a company to buy credits? No. SerpCrypto is a no-KYC SERP API: no ID, no selfie, no company documents. You top up from $25 in BTC, USDT, USDC, ETH, SOL, LTC and buy a pack from your balance. ## Sources 1. [Google num=100 impact data](https://searchengineland.com/google-num100-impact-data-462231) — Search Engine Land (2025-09) --- SerpCrypto: SerpCrypto is a SERP API for developers and AI agents that returns Google and Bing results as JSON, paid in crypto with no ID upload. Prepaid search credits from $0.30/1K searches never expire. No KYC — no ID, no selfie, no documents. Pay in crypto.