Benchmark Radar
AI BENCHMARK PROFILE

BrowseComp (10-agent, prerelease)

General AIKnowledge & Reasoning

BrowseComp accuracy from ten collaborating Opus 5 agents on a pre-release model and unreleased effort configuration.

Released
Unknown
Readiness
Paper only
Primary field
General AI

Primary resources

Benchmark Radar records only publicly supported details and links back to primary sources for verification.