Octen - We stayed quiet and built the FASTEST AI search on Earth
by•
Octen is a search API built for AI agents, not humans. Instead of one query at a time, it decomposes a question into sub-queries, fires them all at once, and returns results in 62ms (P50), several times faster than other providers. Pricing is $1 per 1,000 calls, well below typical rates of $4 to $9. A single account handles over 1,000,000 queries per second, with new content indexed within 5 minutes. Our embedding models rank #1 on RTEB and MMEB-v2.
Replies
Congrats! Octen nails the three things that actually matter in production: cost, latency, and stability. Sub-100ms search keeps real-time copilots and multi-search agent loops feeling instant, the pricing makes high-volume usage genuinely viable instead of something you have to ration, and latency stays flat at 1M+ QPS — no cliff under load, predictable P99 you can build an SLA on. Fast, affordable, and rock-solid all at once is a rare combo. Excited to see where this goes!
As algorithm engineers at Octen, we are committed to delivering the best search results to you as quickly as possible.
Try using Octen in your product, you'll be amazed!
What stands out is the combination of speed, cost, and breadth. Broad research usually means fanning out across many queries and sources, which gets slow and expensive quickly. Octen seems purpose-built for that workload, and the $1/1k pricing makes it practical to run at agent scale.
Octen is exactly the infrastructure AI agents have been waiting for. Instead of forcing the human "one query at a time" habit onto machines, it works the way agents actually think—breaking a question into sub-queries and running them in parallel.
The numbers speak for themselves: 62ms P50 latency (several times faster than anything I've used), $1 per 1,000 calls (vs. the usual $4–$9), 1M+ queries per second on a single account, and 5-minute index freshness. Add embedding models ranked #1 on both RTEB and MMEB-v2, and this stops being "just another search API"—it's a real foundation to build on.
AVI by True AI
65ms is the part that really caught my attention. In something like a voice agent or a multi-agent workflow where one agent is constantly handing off to another, latency compounds very quickly. Keeping each call this fast could make the difference between a demo that looks good and an app that actually feels usable. Definitely trying this with my current app! Congrats on the launch!!
I believe every model will ultimately pursue efficiency. This is essential for user experience, and Octen is on the right track.
Most search APIs do one query at a time, like a human typing in a box. Octen's a different architecture — agents can run many queries at once. Not just faster, built for it.