Perplexity Fast Search cuts per-task cost 68% on six agent benchmarks at 160ms latency
perplexity_ai · x · 2026-09-25
Perplexity shared benchmark numbers for Fast Search:
- Across six agent benchmarks, it cut cost per task by 68% vs the default preset with comparable quality
- 160ms median latency — the lowest among the search APIs they compared
A very concrete cost argument for retrieval-heavy agent applications.
Related event: Perplexity Launches Fast Search Powered by In-House Photon Engine(5 posts)→
More from coding & agent
- Cursor mulls killing Plan Mode for a Shift+Tab effort-level hotkey — steipete · 2026-09-25
- Dev burns 5-10B tokens a day running 24 Devin agents plus Codex and Claude Code — teropa · 2026-09-25
- Question's Gambit tops BrowseComp-Plus recall with 96.6% using only BM25 — CShorten30 · 2026-09-25
- Engineer tests now bait AI tools with plausible-but-wrong answers to grade verification skills — l4rz · 2026-09-25
- AutoScientists NeurIPS paper: self-organizing AI research teams hit 74.4 percentile on BioML-Bench — marinkazitnik · 2026-09-25
- The PM role is now 3 jobs: AI dissolves barriers between customer, engineering, and business — aakashgupta · 2026-09-25