multimodal benchmark
MuirBench Leaderboard
A comprehensive benchmark for robust multi-image understanding capabilities of multimodal LLMs. Consists of 12 diverse multi-image tasks involving 10 categories of multi-image relations (e.g., multiview, temporal relations, narrative, complementary). Comprises 11,264 images and 2,600 multiple-choice questions created in a pairwise manner, where each standard instance is paired with an unanswerable variant for reliable assessment.
Updated Aug 17, 2026
Models12
Model coverage12
MetricScore
EvidenceB
MuirBench Ranking
Higher score ranks better on this benchmark.
Sort by
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelAC | Score80.3% | Percentile100.0% | Participants12 | EvidenceC | Evaluated |
| Rank02 | ModelAC | Score80.1% | Percentile90.9% | Participants12 | EvidenceC | Evaluated |
| Rank03 | ModelAC | Score77.6% | Percentile81.8% | Participants12 | EvidenceC | Evaluated |
| Rank04 | ModelAC | Score76.8% | Percentile72.7% | Participants12 | EvidenceC | Evaluated |
| Rank05 | ModelAC | Score75.0% | Percentile63.6% | Participants12 | EvidenceC | Evaluated |
| Rank06 | ModelAC | Score72.8% | Percentile54.5% | Participants12 | EvidenceC | Evaluated |
| Rank07 | ModelAC | Score72.8% | Percentile45.5% | Participants12 | EvidenceC | Evaluated |
| Rank08 | ModelAC | Score64.4% | Percentile36.4% | Participants12 | EvidenceC | Evaluated |
| Rank09 | ModelAC | Score63.8% | Percentile27.3% | Participants12 | EvidenceC | Evaluated |
| Rank10 | ModelAC | Score62.9% | Percentile18.2% | Participants12 | EvidenceC | Evaluated |
| Rank11 | ModelAC | Score59.2% | Percentile9.1% | Participants12 | EvidenceC | Evaluated |
| Rank12 | ModelLA | Score58.3% | Percentile0.0% | Participants12 | EvidenceC | Evaluated |
MuirBench Highlights
The leading models and scores on this benchmark.
MuirBench Score Distribution
A closer view of the leading scores on this benchmark.
The Top AI Models for MuirBench
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
Ranking basisThis muirbench 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 VL 32B ThinkingAlibaba Cloud / Qwen Team
- Score
- 80.3%
Strengths
- Ranks #1 of 12 compared models
- 100th percentile on this benchmark
- C evidence result
Considerations
- This result measures MuirBench, not total model capability
- 02ACQwen3 VL 235B A22B ThinkingAlibaba Cloud / Qwen Team
- Score
- 80.1%
Strengths
- Ranks #2 of 12 compared models
- 91th percentile on this benchmark
- C evidence result
Considerations
- This result measures MuirBench, not total model capability
- 03ACQwen3 VL 30B A3B ThinkingAlibaba Cloud / Qwen Team
- Score
- 77.6%
Strengths
- Ranks #3 of 12 compared models
- 82th percentile on this benchmark
- C evidence result
Considerations
- This result measures MuirBench, not total model capability
- 04ACQwen3 VL 8B ThinkingAlibaba Cloud / Qwen Team
- Score
- 76.8%
Strengths
- Ranks #4 of 12 compared models
- 73th percentile on this benchmark
- C evidence result
Considerations
- This result measures MuirBench, not total model capability
- 05ACQwen3 VL 4B ThinkingAlibaba Cloud / Qwen Team
- Score
- 75.0%
Strengths
- Ranks #5 of 12 compared models
- 64th percentile on this benchmark
- C evidence result
Considerations
- This result measures MuirBench, not total model capability
Selection summary
Best AI Models for MuirBench
Qwen3 VL 32B Thinking currently leads MuirBench with 80.3%. 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 MuirBench?
What MuirBench measures and how its scores work.
A comprehensive benchmark for robust multi-image understanding capabilities of multimodal LLMs. Consists of 12 diverse multi-image tasks involving 10 categories of multi-image relations (e.g., multiview, temporal relations, narrative, complementary). Comprises 11,264 images and 2,600 multiple-choice questions created in a pairwise manner, where each standard instance is paired with an unanswerable variant for reliable assessment.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- MuirBench
- Modality
- multimodal
- Primary category
- multimodal
- Score direction
- higher
- LLMBoard eligible
- Yes
- Evaluation key
- muirbench|llm-stats-current
Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about MuirBench.
Which model scores highest on MuirBench?
Qwen3 VL 32B Thinking is currently ranked first with 80.3%.
What does MuirBench measure?
A comprehensive benchmark for robust multi-image understanding capabilities of multimodal LLMs. Consists of 12 diverse multi-image tasks involving 10 categories of multi-image relations (e.g., multiview, temporal relations, narrative, complementary). Comprises 11,264 images and 2,600 multiple-choice questions created in a pairwise manner, where each standard instance is paired with an unanswerable variant for reliable assessment.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
12 model results are currently shown.
Does this benchmark affect the overall score?
Yes. This benchmark can contribute to the current LLMBoard capability score.