multimodal benchmark
MMBench-V1.1 Leaderboard
Version 1.1 of MMBench, an improved bilingual benchmark for assessing multi-modal capabilities of vision-language models through multiple-choice questions in both English and Chinese, providing systematic evaluation across diverse vision-language tasks.
Updated Aug 17, 2026
Models20
Model coverage20
MetricScore
EvidenceB
MMBench-V1.1 Ranking
Higher score ranks better on this benchmark.
Sort by
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelAC | Score92.8% | Percentile100.0% | Participants20 | EvidenceC | Evaluated |
| Rank02 | ModelAC | Score92.8% | Percentile94.7% | Participants20 | EvidenceC | Evaluated |
| Rank03 | ModelAC | Score92.6% | Percentile89.5% | Participants20 | EvidenceC | Evaluated |
| Rank04 | ModelAC | Score92.3% | Percentile84.2% | Participants20 | EvidenceC | Evaluated |
| Rank05 | ModelAC | Score91.5% | Percentile79.0% | Participants20 | EvidenceC | Evaluated |
| Rank06 | ModelAC | Score90.8% | Percentile73.7% | Participants20 | EvidenceC | Evaluated |
| Rank07 | ModelAC | Score90.6% | Percentile68.4% | Participants20 | EvidenceC | Evaluated |
| Rank08 | ModelAC | Score89.9% | Percentile63.2% | Participants20 | EvidenceC | Evaluated |
| Rank09 | ModelAC | Score88.9% | Percentile57.9% | Participants20 | EvidenceC | Evaluated |
| Rank10 | ModelAC | Score87.5% | Percentile52.6% | Participants20 | EvidenceC | Evaluated |
| Rank11 | ModelAC | Score87.0% | Percentile47.4% | Participants20 | EvidenceC | Evaluated |
| Rank12 | ModelAC | Score86.7% | Percentile42.1% | Participants20 | EvidenceC | Evaluated |
| Rank13 | ModelAC | Score85.1% | Percentile36.8% | Participants20 | EvidenceC | Evaluated |
| Rank14 | ModelAC | Score85.0% | Percentile31.6% | Participants20 | EvidenceC | Evaluated |
| Rank15 | ModelAC | Score81.8% | Percentile26.3% | Participants20 | EvidenceC | Evaluated |
| Rank16 | ModelLA | Score81.0% | Percentile21.1% | Participants20 | EvidenceC | Evaluated |
| Rank17 | ModelDE | Score79.3% | Percentile15.8% | Participants20 | EvidenceC | Evaluated |
| Rank18 | ModelDE | Score79.2% | Percentile10.5% | Participants20 | EvidenceC | Evaluated |
| Rank19 | ModelCO | Score68.7% | Percentile5.3% | Participants20 | EvidenceC | Evaluated |
| Rank20 | ModelDE | Score68.3% | Percentile0.0% | Participants20 | EvidenceC | Evaluated |
MMBench-V1.1 Highlights
The leading models and scores on this benchmark.
MMBench-V1.1 Score Distribution
A closer view of the leading scores on this benchmark.
The Top AI Models for MMBench-V1.1
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
Ranking basisThis mmbench-v1.1 AI model leaderboard uses descending score in the benchmark's original unit. The leaderboard ranking keeps matched price and speed data separate from benchmark evidence.
- 01ACQwen3.5-122B-A10BAlibaba Cloud / Qwen Team
- Score
- 92.8%
- Price
- $0.40 input / $3.2 output per 1M tokens
Strengths
- Ranks #1 of 20 compared models
- 100th percentile on this benchmark
- C evidence result
Considerations
- This result measures MMBench-V1.1, not total model capability
- 02ACQwen3.6-35B-A3BAlibaba Cloud / Qwen Team
- Score
- 92.8%
- Price
- $0.25 input / $1.5 output per 1M tokens
Strengths
- Ranks #2 of 20 compared models
- 95th percentile on this benchmark
- C evidence result
Considerations
- This result measures MMBench-V1.1, not total model capability
- 03ACQwen3.5-27BAlibaba Cloud / Qwen Team
- Score
- 92.6%
- Price
- $0.30 input / $2.4 output per 1M tokens
- Speed
- Up to 6.7 tok/s via Novita
Strengths
- Ranks #3 of 20 compared models
- 89th percentile on this benchmark
- C evidence result
Considerations
- This result measures MMBench-V1.1, not total model capability
- 04ACQwen3.6-27BAlibaba Cloud / Qwen Team
- Score
- 92.3%
- Price
- $0.60 input / $3.6 output per 1M tokens
- Speed
- Up to 6.1 tok/s via Novita
Strengths
- Ranks #4 of 20 compared models
- 84th percentile on this benchmark
- C evidence result
Considerations
- This result measures MMBench-V1.1, not total model capability
- 05ACQwen3.5-35B-A3BAlibaba Cloud / Qwen Team
- Score
- 91.5%
- Price
- $0.25 input / $2.0 output per 1M tokens
Strengths
- Ranks #5 of 20 compared models
- 79th percentile on this benchmark
- C evidence result
Considerations
- This result measures MMBench-V1.1, not total model capability
Selection summary
Best AI Models for MMBench-V1.1
Qwen3.5-122B-A10B currently leads MMBench-V1.1 with 92.8%. It is the top model on this specific benchmark, while the best LLM for the broader task should also be checked against other benchmarks, price and runtime.
Use this leaderboard with the supporting benchmark results and coverage details above. A leaderboard position summarizes the selected ranking signal; it does not replace workload-specific testing.
What is MMBench-V1.1?
What MMBench-V1.1 measures and how its scores work.
Version 1.1 of MMBench, an improved bilingual benchmark for assessing multi-modal capabilities of vision-language models through multiple-choice questions in both English and Chinese, providing systematic evaluation across diverse vision-language tasks.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- MMBench-V1.1
- Modality
- multimodal
- Primary category
- multimodal
- Score direction
- higher
- LLMBoard eligible
- Yes
- Evaluation key
- mmbench-v1.1|llm-stats-current
Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about MMBench-V1.1.
Which model scores highest on MMBench-V1.1?
Qwen3.5-122B-A10B is currently ranked first with 92.8%.
What does MMBench-V1.1 measure?
Version 1.1 of MMBench, an improved bilingual benchmark for assessing multi-modal capabilities of vision-language models through multiple-choice questions in both English and Chinese, providing systematic evaluation across diverse vision-language tasks.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
20 model results are currently shown.
Does this benchmark affect the overall score?
Yes. This benchmark can contribute to the current LLMBoard capability score.