multimodal benchmark
MM-BrowserComp
MM-BrowserComp evaluates multimodal agents on web browsing and information retrieval tasks, testing a model's ability to perceive, navigate, and extract information from real web environments.
Updated Aug 11, 2026
Models1
Model coverage1
MetricScore
EvidenceB
MM-BrowserComp Ranking
Higher score ranks better on this benchmark.
| 01 | XI | 52.0% | 100.0% | 1 | C |
MM-BrowserComp Highlights
The leading models and scores on this benchmark.
What is MM-BrowserComp?
What MM-BrowserComp measures and how its scores work.
MM-BrowserComp evaluates multimodal agents on web browsing and information retrieval tasks, testing a model's ability to perceive, navigate, and extract information from real web environments.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- MM-BrowserComp
- Modality
- multimodal
- Primary category
- multimodal
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- mm-browsercomp|llm-stats-current
Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about MM-BrowserComp.
Which model scores highest on MM-BrowserComp?
MiMo-V2-Omni is currently ranked first with 52.0%.
What does MM-BrowserComp measure?
MM-BrowserComp evaluates multimodal agents on web browsing and information retrieval tasks, testing a model's ability to perceive, navigate, and extract information from real web environments.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
1 model results are currently shown.
Does this benchmark affect the overall score?
No. This benchmark is shown for reference but does not contribute to the overall score.