BrowseComp
A benchmark of 1,266 hard-to-find, multi-hop web-browsing questions whose answers are difficult to locate but easy to verify, measuring an agent's ability to persistently search and synthesize information from the web.
What this benchmark measures
A benchmark of 1,266 hard-to-find, multi-hop web-browsing questions whose answers are difficult to locate but easy to verify, measuring an agent's ability to persistently search and synthesize information from the web.
Rows on this page are sourced from public benchmark artifacts, leaderboard exports, or source-linked model reports. Each row keeps benchmark version, source model name, and available run details attached to the score.
The metric shown here is accuracy. It should be interpreted within BrowseComp, not compared as part of a site-wide ranking.
Frequently asked
What is BrowseComp?
A benchmark of 1,266 hard-to-find, multi-hop web-browsing questions whose answers are difficult to locate but easy to verify, measuring an agent's ability to persistently search and synthesize information from the web. It is a agents benchmark measured by accuracy.
What does accuracy mean on BrowseComp?
BrowseComp reports accuracy (%); higher is better. Scores are shown only within BrowseComp and are never averaged with other benchmarks.
What is the top reported BrowseComp score?
GPT-5.6 Sol Ultra has the top reported score on BrowseComp: 92.2% (accuracy).
Why do BrowseComp scores differ across runs?
Harness, scaffold, reasoning effort, and prompt setup change results, so two runs of the same model can differ. evals.report keeps each score with its run context so the differences stay visible.
Does evals.report rank models across benchmarks?
No. BrowseComp scores are shown within their own metric; evals.report never combines benchmarks into a composite ranking or a single "best model".