Versions

4 versions · Metric: BrowseComp Accuracy

Versions of BrowseComp
VersionReleasedModelsResults
BrowseComp With context management — Inkling evaluationConfiguration: Context management (Inkling evaluation)July 15, 202622
BrowseComp With context management — MiniMax M2.5Configuration: Context management (MiniMax M2.5)February 12, 202611
BrowseComp 130-question subset — Mercor web research agentConfiguration: Mercor web research agent, 100 steps, 3 hoursApril 10, 20253536
BrowseCompLatestApril 10, 20252027

Recently reported results: BrowseComp

Latest version: BrowseComp