OpenSERP is a free, open-source SERP API and CLI that returns live, browser-rendered search results from Google, Bing, Yandex, Baidu, DuckDuckGo, and Ecosia — engines and coverage that paid search APIs often miss. It has about 1,500 stars on GitHub, is actively developed in 2026, and is built to feed search results to LLMs, agents, and SEO tooling.
Most search APIs charge per query and require API keys; OpenSERP takes the self-hosted route. One command or a prebuilt Docker image gives you a local service that queries six engines through a real browser and returns structured results in a single consistent JSON schema, including SERP features like AI summaries, answer boxes, people-also-ask, and related searches. Its Megasearch mode runs one query across several engines at once and merges and dedupes the results, and an extraction option returns clean markdown of the target pages alongside the results — exactly what retrieval pipelines and research agents need. It also supports image search, language and date filters, proxies, caching, and a resilient mode for automation at scale.
OpenSERP is free and open source under the MIT license for self-hosting. If you would rather not run the infrastructure yourself, the project also offers a hosted cloud version with the same API, which is the paid route.
Developers building LLM agents, RAG pipelines, and research tools that need real web search, plus SEO practitioners who want rank-tracking data — including from Yandex and Baidu — without per-query API bills.