Octen Search Tops AI Agent Search Benchmark at 0.21s Per Query, Costs 1/7 of Exa
kimmonismus · x · 2026-09-25
Artificial Analysis' Search API benchmark shows Octen delivers live web search for AI agents at 0.21 seconds per query — the only Search API ranking top-3 on quality, speed, and cost. Search fees run $9.07 per 1,000 tasks vs. $65.57 for Exa auto (roughly one-seventh); with model costs, totals are $58.22 vs. $127.15. The benchmark keeps the answer model fixed across DeepSearchQA, BrowseComp and AA-Omniscience, isolating how search providers affect agent answers. Octen Search (highlights) scores 77, tied for third, and leads speed at 15.6 seconds per task. Its Broad Search fans one query into parallel searches, and its Deep Research product claims cited reports in 2-3 minutes. Data as of September 8.
More from coding & agent
- Demo: leave comments on a paper and let your agent address them — Rasmic · 2026-09-25
- Jev becomes fastest-adopted AI model ever, spawning sub-cent browser and trading agents in 3 days — multiply_matrix · 2026-09-25
- Making Games for Agents Means You Have Instant AI Playtesters — One Prompt Upgraded the Graphics — tristanbob · 2026-09-25
- DeepLedger Launches AI Staff Accountant MCP Connector for QuickBooks — modelcontextprotocol · 2026-09-25
- AgentWallet MCP Gives AI Agents Wallet Infrastructure Across Solana and EVM Chains — modelcontextprotocol · 2026-09-25
- New SDK Tests AI Agents on Real Websites and Workflows, Not Mocked Tools — Able-Ingenuity8885 · 2026-09-25