multimodal benchmark
MMMU Leaderboard
MMMU is a multimodal benchmark evaluating models on college-level subject knowledge and deliberate reasoning across 30 subjects and 183 subfields.
Updated Sep 5, 2026
Models64
Model coverage64
MetricScore
EvidenceB
MMMU Ranking
Higher score ranks better on this benchmark.
Sort by
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelAC | Score86.00% | Percentile100.00% | Participants64 | EvidenceC | Evaluated |
| Rank02 | ModelOP | Score85.40% | Percentile98.41% | Participants64 | EvidenceC | Evaluated |
| Rank03 | ModelOP | Score85.40% | Percentile96.83% | Participants64 | EvidenceC | Evaluated |
| Rank04 | ModelOP | Score85.40% | Percentile95.24% | Participants64 | EvidenceC | Evaluated |
| Rank05 | ModelOP | Score84.20% | Percentile93.65% | Participants64 | EvidenceC | Evaluated |
| Rank06 | ModelAC | Score83.90% | Percentile92.06% | Participants64 | EvidenceC | Evaluated |
| Rank07 | ModelOP | Score82.90% | Percentile90.48% | Participants64 | EvidenceC | Evaluated |
| Rank08 | ModelAC | Score82.90% | Percentile88.89% | Participants64 | EvidenceC | Evaluated |
| Rank09 | ModelAC | Score82.30% | Percentile87.30% | Participants64 | EvidenceC | Evaluated |
| Rank10 | ModelGO | Score82.00% | Percentile85.71% | Participants64 | EvidenceC | Evaluated |
| Rank11 | ModelAC | Score81.70% | Percentile84.13% | Participants64 | EvidenceC | Evaluated |
| Rank12 | ModelOP | Score81.60% | Percentile82.54% | Participants64 | EvidenceC | Evaluated |
| Rank13 | ModelAC | Score81.40% | Percentile80.95% | Participants64 | EvidenceC | Evaluated |
| Rank14 | ModelGO | Score79.70% | Percentile79.37% | Participants64 | EvidenceC | Evaluated |
| Rank15 | ModelGO | Score79.60% | Percentile77.78% | Participants64 | EvidenceC | Evaluated |
| Rank16 | ModelLA | Score78.70% | Percentile76.19% | Participants64 | EvidenceC | Evaluated |
| Rank17 | ModelST | Score78.10% | Percentile74.60% | Participants64 | EvidenceC | Evaluated |
| Rank18 | ModelXA | Score78.00% | Percentile73.02% | Participants64 | EvidenceC | Evaluated |
| Rank19 | ModelOP | Score77.60% | Percentile71.43% | Participants64 | EvidenceC | Evaluated |
| Rank20 | ModelGO | Score75.40% | Percentile69.84% | Participants64 | EvidenceC | Evaluated |
| Rank21 | ModelOP | Score75.20% | Percentile68.25% | Participants64 | EvidenceC | Evaluated |
| Rank22 | ModelCO | Score75.10% | Percentile66.67% | Participants64 | EvidenceC | Evaluated |
| Rank23 | ModelAN | Score75.00% | Percentile65.08% | Participants64 | EvidenceC | Evaluated |
| Rank24 | ModelOP | Score74.80% | Percentile63.49% | Participants64 | EvidenceC | Evaluated |
| Rank25 | ModelAN | Score74.40% | Percentile61.90% | Participants64 | EvidenceC | Evaluated |
| Rank26 | ModelME | Score73.40% | Percentile60.32% | Participants64 | EvidenceC | Evaluated |
| Rank27 | ModelGO | Score72.90% | Percentile58.73% | Participants64 | EvidenceC | Evaluated |
| Rank28 | ModelOP | Score72.70% | Percentile57.14% | Participants64 | EvidenceC | Evaluated |
| Rank29 | ModelOP | Score72.20% | Percentile55.56% | Participants64 | EvidenceC | Evaluated |
| Rank30 | ModelGO | Score70.70% | Percentile53.97% | Participants64 | EvidenceC | Evaluated |
MMMU Highlights
The leading models and scores on this benchmark.
MMMU Score Distribution
A closer view of the leading scores on this benchmark.
The Top AI Models for MMMU
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
What is MMMU?
What MMMU measures and how its scores work.
MMMU (Massive Multi-discipline Multimodal Understanding) is a benchmark containing 11.5K multimodal questions from college exams, quizzes, and textbooks across six core disciplines: Art & Design, Business, Science, Health & Medicine, Humanities & Social Science, and Tech & Engineering.
It measures multimodal models' college-level subject knowledge and deliberate reasoning using Score, reported as a ratio.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- MMMU
- 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.
Which model scores highest on MMMU?
Qwen3.6 Plus is currently ranked first with 86.00%.
What are the top three models on MMMU?
The current leaders are Qwen3.6 Plus (86.00%), GPT-5.1 (85.40%), and GPT-5.1 Instant (85.40%).
Which MMMU model has the lowest official input price?
Nova Lite has the lowest matched official input price at $0.06 input / $0.24 output per 1M tokens.
Which models are fastest among MMMU results?
The fastest matched records are Llama 4 Scout (776.10 tok/s via Groq), GPT-4.1 mini (467.39 tok/s via OpenAI), and Llama 4 Maverick (307.30 tok/s via Groq).
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.