multimodal benchmark
MobileMiniWob++_SR
MobileMiniWob++ SR (Success Rate) is an adaptation of the MiniWob++ web interaction benchmark for mobile Android environments within AndroidWorld. It comprises 92 web interaction tasks adapted for touch-based mobile interfaces, evaluating agents' ability to navigate and interact with web applications on mobile devices.
Updated Aug 11, 2026
Models2
Model coverage2
MetricScore
EvidenceB
MobileMiniWob++_SR Ranking
Higher score ranks better on this benchmark.
| 01 | AC | 91.4% | 100.0% | 2 | C | |
| 02 | AC | 68.0% | 0.0% | 2 | C |
MobileMiniWob++_SR Score Distribution
A closer view of the leading scores on this benchmark.
MobileMiniWob++_SR Highlights
The leading models and scores on this benchmark.
What is MobileMiniWob++_SR?
What MobileMiniWob++_SR measures and how its scores work.
MobileMiniWob++ SR (Success Rate) is an adaptation of the MiniWob++ web interaction benchmark for mobile Android environments within AndroidWorld. It comprises 92 web interaction tasks adapted for touch-based mobile interfaces, evaluating agents' ability to navigate and interact with web applications on mobile devices.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- MobileMiniWob++_SR
- Modality
- multimodal
- Primary category
- multimodal
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- mobileminiwob++-sr|llm-stats-current
Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about MobileMiniWob++_SR.
Which model scores highest on MobileMiniWob++_SR?
Qwen2.5 VL 7B Instruct is currently ranked first with 91.4%.
What does MobileMiniWob++_SR measure?
MobileMiniWob++ SR (Success Rate) is an adaptation of the MiniWob++ web interaction benchmark for mobile Android environments within AndroidWorld. It comprises 92 web interaction tasks adapted for touch-based mobile interfaces, evaluating agents' ability to navigate and interact with web applications on mobile devices.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
2 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.