Benchmarks/agents/BrowseComp-VL

BrowseComp-VL

BrowseComp-VL is the vision-language variant of BrowseComp, evaluating multimodal models on web browsing comprehension tasks that require processing visual web page content alongside text.

Progress Over Time

Interactive timeline showing model performance evolution on BrowseComp-VL

State-of-the-art frontier
Open
Proprietary

BrowseComp-VL Leaderboard

1 models
ContextCostLicense
1
Zhipu AI
GLM-5V-Turbo
Zhipu AI
———
Notice missing or incorrect data?

FAQ

Common questions about BrowseComp-VL

BrowseComp-VL is the vision-language variant of BrowseComp, evaluating multimodal models on web browsing comprehension tasks that require processing visual web page content alongside text.
The BrowseComp-VL leaderboard ranks 1 AI models based on their performance on this benchmark. Currently, GLM-5V-Turbo by Zhipu AI leads with a score of 0.519. The average score across all models is 0.519.
The highest BrowseComp-VL score is 0.519, achieved by GLM-5V-Turbo from Zhipu AI.
1 models have been evaluated on the BrowseComp-VL benchmark, with 0 verified results and 1 self-reported results.
BrowseComp-VL is categorized under agents, multimodal, search, and vision. The benchmark evaluates multimodal models.