For SEO & rank tracking
Search every engine at once
See where you rank across Google, Bing, Yandex, Baidu, DuckDuckGo, and Ecosia from a single query - no engine-by-engine scraping.
GET /v1/mega/search?text=keyword&mode=balancedThe managed path for the open-source OpenSERP engine. Send a query, get structured results back. No browsers, proxies, or retries to operate.
Start free, then pay as you go: $0.009 per 10 returned web results. Errors are free, no subscription. See pricing →
No subscription
No minimum
Credits never expire
Errors are free
OpenSERP Cloud is strongest when one search needs more than a single Google result page can give you.
For SEO & rank tracking
See where you rank across Google, Bing, Yandex, Baidu, DuckDuckGo, and Ecosia from a single query - no engine-by-engine scraping.
GET /v1/mega/search?text=keyword&mode=balancedFor AI & RAG pipelines
A page that ranks across several independent engines is far less likely to be a fake or a hallucination. Keep those agreed-on sources to ground your LLM answers.
results.filter(r => r.engine_count >= 2) // agreed across enginesSend a query with your API key and get structured JSON back. Credits used and remaining are returned in response headers.
curl -H 'Authorization: Bearer osk_live_xxx' \
'https://api.openserp.org/v1/google/search?text=open+source+serp+api&limit=10'{
"meta": {
"engines_failed": [],
"request_id": "019eb88e-d8ec-7c66-b50a-b8ff662a9db6",
"requested_at": "2026-06-11T21:20:29Z",
"took_ms": 1201,
"version": "2.1"
},
"pagination": {
"has_more": true,
"next_start": 10,
"page": 1
},
"query": {
"engines_requested": [
"google"
],
"text": "open source serp api"
},
"results": [
{
"classification": {
"content_type": "webpage",
"source_hint": "code_repository"
},
"display_url": "github.com › karust › openserp",
"domain": "github.com",
"domain_info": {
"category": "",
"sld": "github",
"tld": "com"
},
"engine": "google",
"favicon": "https://github.com/favicon.ico",
"id": "s_c84e44351617d71f",
"position": {
"absolute": 1
},
"rank": 1,
"snippet": "OpenSERP is a free, open-source API and CLI for accessing normalized search engine results from Google, Yandex, Baidu, Bing, DuckDuckGo, and Ecosia.",
"title": "karust/openserp: Open-source SERP API for AI, SEO & ...",
"type": "organic",
"url": "https://github.com/karust/openserp"
},
{
"classification": {
"content_type": "webpage",
"source_hint": "social_forum"
},
"display_url": "reddit.com › r › LocalLLaMA › comments › 1c62wmz › ...",
"domain": "reddit.com",
"domain_info": {
"category": "forum",
"sld": "reddit",
"tld": "com"
},
"engine": "google",
"favicon": "https://reddit.com/favicon.ico",
"id": "s_4e18eaf95d083f7e",
"position": {
"absolute": 2
},
"rank": 2,
"snippet": "They're not entirely free but at least they offer a solid free plan for Google SERP scraping. You get 1,000 free API credits with Scrapingbee.",
"title": "Serp Alternative: Google result scrape- Hopefully free",
"type": "organic",
"url": "https://www.reddit.com/r/LocalLLaMA/comments/1c62wmz/serp_alternative_google_result_scrape_hopefully"
}
],
"serp_features": []
}OpenSERP already handles browser automation, multi-engine parsing, and retries. Cloud runs the hosted infrastructure so your team can use the API directly.
Call one REST endpoint and get structured search results back. No browsers to deploy, no infrastructure to keep alive.
Cloud is the hosted path for the same open-source engine you can already run yourself. Same coverage and quality, without the operational work.
Single-engine, image, and megasearch endpoints across Google, Bing, Yandex, Baidu, DuckDuckGo, and Ecosia, plus SDKs, an MCP server, and an n8n node.
Use OpenSERP Cloud for product features, SEO campaigns, research jobs, and AI pipelines that need live search data.
Integrate live SERP data into apps, dashboards, or AI agents with a small REST API. No scraper deployment, no maintenance burden.
Track keyword rankings across Google, Yandex, Baidu, and Bing, then monitor competitor listings without provisioning your own scraping infrastructure.
Ground LLM responses and retrieval workflows on fresh, multi-engine search results. Call it directly, via the SDKs, or through the MCP server.
OSS is production-ready right now and free forever. Cloud is the managed path when you do not want to operate the infrastructure yourself.
Free · MIT License
Managed · Free credits to start
Create an account from any page on this site. Activation takes minutes and includes free starter credits. No card required.
Yes. Your invite comes with free starter credits so you can run real searches right away. Describe your use case when you join to start with more.
It uses the same OpenSERP engine, fully managed by us. You call an API and get structured search results back. No browsers to deploy and no infrastructure to keep alive.
Pay-as-you-go credits: $0.009 per credit. Web search uses 1 credit per 10 returned result positions, rounded up. Higher-cost operations like images, extraction, and balanced megasearch are shown before you run them. Errors are free. No subscription.
Yes. The OSS version is production-ready and free under MIT. Run it with Docker or Go on your own infrastructure. Cloud is the managed path and starts with free credits.
Google, Bing, Yandex, Baidu, DuckDuckGo, and Ecosia from a single endpoint. The same engines as OpenSERP OSS, without the infrastructure.