Audience
AI labs, agent builders, and developers seeking to give AI systems fast, scalable access to current and historical web information
About Keenable
Keenable is independent web search infrastructure for AI labs, inference platforms and agents. Crawler, index and ranking run in one stack, over 100B+ web documents updated in real time.
The Search API gives agents two tools: search_web_pages finds pages, fetch_page_content returns a page as clean markdown. Both work through a REST API, an MCP server and a CLI on one API key. Latency is under 250ms p95 in US East.
Search quality is tracked on NEEDLE, Keenable's public benchmark. It reruns daily, sends the same queries to several search providers and scores each against an oracle ranking built from their pooled results.
For high-volume workloads, the Frontier Tier adds dedicated capacity, pricing from $1 per 1,000 requests at 100 RPS and above, and deployment in the cloud or on-premises.
Time Machine, in early access, adds point-in-time search gated on Keenable's own crawl date: set query_time and both the corpus and the ranking rewind to that moment.
Pricing
For AI Labs and Inference platforms: $1 / 1000 requests at 100 RPS+