multimodal benchmark
MMMU (val) Leaderboard
MMMU (val) is the validation set of the Massive Multi-discipline Multimodal Understanding and Reasoning benchmark, covering college-level multimodal questions across 6 core disciplines, 30 subjects, and 183 subfields.
Updated Sep 5, 2026
Models13
Model coverage13
MetricScore
EvidenceB
MMMU (val) Ranking
Higher score ranks better on this benchmark.
Sort by
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelAC | Score78.10% | Percentile100.00% | Participants13 | EvidenceC | Evaluated |
| Rank02 | ModelAC | Score76.00% | Percentile91.67% | Participants13 | EvidenceC | Evaluated |
| Rank03 | ModelAC | Score76.00% | Percentile83.33% | Participants13 | EvidenceC | Evaluated |
| Rank04 | ModelAC | Score74.20% | Percentile75.00% | Participants13 | EvidenceC | Evaluated |
| Rank05 | ModelAC | Score74.10% | Percentile66.67% | Participants13 | EvidenceC | Evaluated |
| Rank06 | ModelAC | Score70.80% | Percentile58.33% | Participants13 | EvidenceC | Evaluated |
| Rank07 | ModelAC | Score69.60% | Percentile50.00% | Participants13 | EvidenceC | Evaluated |
| Rank08 | ModelAC | Score67.40% | Percentile41.67% | Participants13 | EvidenceC | Evaluated |
| Rank09 | ModelGO | Score64.90% | Percentile33.33% | Participants13 | EvidenceC | Evaluated |
| Rank10 | ModelGO | Score59.60% | Percentile25.00% | Participants13 | EvidenceC | Evaluated |
| Rank11 | ModelGO | Score48.80% | Percentile16.67% | Participants13 | EvidenceC | Evaluated |
| Rank12 | ModelLA | Score48.40% | Percentile8.33% | Participants13 | EvidenceC | Evaluated |
| Rank13 | ModelCO | Score32.90% | Percentile0.00% | Participants13 | EvidenceC | Evaluated |
MMMU (val) Highlights
The leading models and scores on this benchmark.
MMMU (val) Score Distribution
A closer view of the leading scores on this benchmark.
The Top AI Models for MMMU (val)
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
What is MMMU (val)?
What MMMU (val) measures and how its scores work.
A validation set of the Massive Multi-discipline Multimodal Understanding and Reasoning benchmark with questions using diverse image types, including charts, diagrams, maps, and tables.
It reports Score as a ratio.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- MMMU (val)
- Modality
- multimodal
- Primary category
- multimodal
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- overall
LLM Stats. Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about MMMU (val).
Which model scores highest on MMMU (val)?
Qwen3 VL 32B Thinking is currently ranked first with 78.10%.
What are the top three models on MMMU (val)?
The current leaders are Qwen3 VL 32B Thinking (78.10%), Qwen3 VL 30B A3B Thinking (76.00%), and Qwen3 VL 32B Instruct (76.00%).
Which MMMU (val) model has the lowest official input price?
No matched official input price is currently available.
Which models are fastest among MMMU (val) results?
The fastest matched records are Gemma 3 27B (33.00 tok/s via DeepInfra), Gemma 3 12B (33.00 tok/s via DeepInfra), and Gemma 3 4B (33.00 tok/s via DeepInfra).
Does the highest score result prove overall model quality?
No. This benchmark measures one defined capability or task. The overall LLMBoard score uses a separate aggregation across eligible benchmark evidence.