BrowseComp-Plus

Evaluates research agents against a fixed document corpus, separately from live-web BrowseComp.

1 models

Higher is better.

Higher is better. Bar colour: blue proprietary, orange open weights, green other.

Authors: Zijian Chen et al. (2025)
Source: BrowseComp-Plus: A More Fair and Transparent Evaluation Benchmark of Deep-Research Agent ↗

Scores are as reported by model developers or evaluators and may use different settings (for example reasoning effort or tools). Compare with care.