Octen debuts #3 on Artificial Analysis Search Index, fastest at 16.9s per task
Scobleizer · x · 2026-09-11
AI search infrastructure startup Octen has debuted on Artificial Analysis's independent Search API benchmark, scoring 77 and ranking third behind only Perplexity Search (medium, 80) and Perplexity Search (high, 79).
Highlights:
- #1 in speed: fastest Time per Task at 16.9 seconds
- #2 in cost efficiency
- Top 3 in search quality, led by its BrowseComp performance
- The only major provider in Artificial Analysis's "Most Attractive Quadrant" for quality vs. cost
Octen's thesis: AI agents don't search like humans — they fire hundreds of parallel queries, read everything retrieved, and need clean, grounded content back in milliseconds at scale-friendly costs. Rather than adapting human search, the team rebuilt it from first principles. The benchmark pinned a fixed base model inside Artificial Analysis's open-source agent harness, Stirrup, changing only the search tool provider.
Related event: Octen Search Ranks Third in Benchmark with Fastest Speed, Lowest Cost(3 posts)→
More from coding & agent
- HABIT-Bench: Best agent memory system scores just 33.61% at inferring user habits — SJTU · 2026-10-28
- Memory Tax study: dedicated agent memory systems often lose to naive RAG — UIUC-CS · 2026-10-28
- Regehr's 'Piano Test': would you trust formally verified code with your life? — moyix · 2026-10-28
- Vibe-coded 'AI 浮生记' game satirizes AI startup life with compute black-market trading — vista8 · 2026-10-28
- Cursor ships Annotate in design mode: agents take visual prompts via on-page markup — soleio · 2026-10-28
- The legendary fake Go generics: Unicode Canadian Syllabics as angle brackets — rickasaurus · 2026-10-28