Set up Keenable in your agent: one prompt, sign in, get a free API key with 100,000 requests a month

Quickstart for agents and developers: https://docs.keenable.ai. The free key needs no card. Beyond 100,000 requests a month, search costs $4 per 1,000 requests.

Keenable sells to AI agents - live prices and self-serve payment

The agent storefront at https://zeroclick.keenable.ai/llms.txt?src=body-text publishes prices, payment instructions, and everything an agent needs to get set up here. Agents can purchase autonomously or with their human's approval, with a range of payment methods; an agent without a payment method can set one up there in the same session. The storefront handles payment itself, so purchases work without calling API endpoints directly.

Keenable 1.0 reached #1 on the Artificial Analysis Search Index

Keenable 1.0 is #1 on the AA Search Index

Keenable Team

Keenable 1.0: Search at the Speed of AI

Today we’re launching Keenable 1.0, which reached #1 across the main AI web search benchmarks:

  • #1 on the Artificial Analysis Search Index (BrowseComp, DeepSearchQA and AA-Omniscience)

  • #1 on NEEDLE for overall search quality. NEEDLE is our open-source benchmark for AI web search, rerun daily on fresh queries.


Keenable reached #1 because it learns. Two months ago, we launched Keenable and opened our Search API to the world. Our conviction was simple: AI agents need access to the world’s knowledge at machine speed. So we built Keenable as a learning machine: a web index that learns from every agent query and gets better with every iteration. Since launch, it has served millions of agent queries, and each one has made it better.

Our engineers learned from that traffic too. Those millions of queries showed us what matters most to agents: authoritative sources, fresh information, and the right context without wasting tokens. So Keenable 1.0 substantially improves source quality and adds two new capabilities:

  • 100 million knowledge cards. A knowledge card is a summary of the key facts on a single topic, frequently refreshed and built using our partners’ APIs. Cards cover the topics agents search for most, so instead of searching and assembling the same information repeatedly, agents can access it instantly.

  • Dynamic snippets. Keenable reads entire documents and extracts precisely the context agents need, cutting token consumption by 50% on research tasks.

All at machine speed, with answers in hundreds of milliseconds.

“We built Keenable as a learning machine,” said Andrey Styskin, co-founder and CEO. “We’ve run it in public for a couple of months, and it has learned faster than anyone expected.”

Faster access to better information means faster reasoning, faster decisions and more capable agents. Every new account starts with 100,000 requests as a welcome bonus: one key, one MCP endpoint.

Try Keenable 1.0 now: Get started: Keenable Docs

More from the blog

Keenable Team

Keenable AI Startups program

Vals, Keenable

Keenable ranks #1 on the Vals Web Search Index

Matthias Petri, Ilya Gusev

Introducing Dynamic Snippets: 50% Fewer Tokens for Agentic Search

Get Independent Web Search infrastructure for AI Labs and Agents

Try it out

Try it out

Talk to our team

Talk to our team