Keenable AI

@KeenableAI

Building web search infrastructure for AI labs and agents

San Francisco, California

keenable.ai

Joined February 2026

148 Following

1,919 Followers

Replies Reposts Media

Pinned
Keenable AI
@KeenableAI
10h
Keenable Search API takes #1 spot on Vals Web Search Index On GPT-5.6 Sol, Keenable pushes the quality frontier from Exa’s 45.24% to 49.08%, at less than half the cost per task Get started with Keenable, first 100K requests on us
Vals AI
@ValsAI
10h
Keenable (Pro mode) is now #1 on the Vals Web Search Index, scoring 49.1% overall. We tested Keenable’s search API paired with GPT-5.6 Sol. It leads in finance at 53.9% and places second in legal at 44.2%, behind Claude Fable 5 with native search.
3
Keenable AI
@KeenableAI
Oct 15
Introducing Keenable Dynamic Snippets Now your agent can spend 50% fewer tokens on search tasks and look deeper into documents. Enabled for everyone. Read our technical report to see how we built Dynamic Snippets while preserving our 200ms p95 search latency:
00:00
3
Keenable AI
@KeenableAI
Oct 6
Your Named Agents can sign up for Keenable and get the web search infrastructure they actually need
Haakam Aujla
@haakamaujla
Oct 6
Introducing AgentID by @agentmail, "Sign in with Google" for AI Agents Millions of agents already use AgentMail. AgentID lets them sign in to your app. Integrate with the prompt below, reply with a screenshot, and we'll give you 3 months free of any AgentMail plan.
00:00
4
Keenable AI
@KeenableAI
Oct 6
If you also don't remember the exact title of the tutorials, you can find them with Keenable in your Claude
00:00
12
Keenable AI
@KeenableAI
Oct 2
Love seeing more work on better search evals for agents. @ValsAI uses private, expert-written tasks. With NEEDLE, we took a different approach: fresh queries every run, with the fully public benchmark.
Vals AI
@ValsAI
Oct 2
We realized that web search wasn’t being measured correctly and we weren’t alone. Other companies have created their own benchmarks to tackle this problem, but kept running into the same issue: data contamination, realism, and answer leakage. When benchmarks measure the wrong