Perplexity: agent grep exploration hits 0.627 answer accuracy vs ~0.37 for semantic search
perplexity_ai · x · 2026-10-13
In its Projects architecture series, Perplexity shares an internal benchmark: on 150 financial-document questions, an exploration agent using grep and standard file tools on a real checkout reached 0.627 answer accuracy versus 0.37 for semantic search, with retrieval scores of 0.867 vs 0.39 — because sandboxes hold real checkouts that standard file tools can search directly.
Related event: Perplexity Explains Projects: A Git-Based Shared File System for Agents(6 posts)→
More from coding & agent
- FactoryAI ships Environments, cloud sandbox for agents, now GA — matanSF · 2026-10-13
- Researcher on doing science in the agent era: ask the hard questions, verify every line — CatAstro_Piyush · 2026-10-13
- "Give me an LLM and a test suite": Archimedes quote gets an AI rewrite — maxleiter · 2026-10-13
- Vibe-coded port of a language itself cuts typechecking time 3x with Opus — maxleiter · 2026-10-13
- Network engineer of 20 years: once you go declarative for AI automation, you never go back — tristanbob · 2026-10-13
- Muse Wonder scores ~42% on senior engineer benchmark vs Sol 6.1's 71% — danshipper · 2026-10-13