llmboard.aiAI model intelligence
Home

Model Rankings

OverallOpen ModelsAgentCodingReasoningMathKnowledgeInstruction FollowingTextVision
Image GenerationImage Editing
Video GenerationImage to VideoVideo Editing
Text to SpeechSpeech to Text
Embeddings

Efficiency

Chat Token PricingImage PricingVideo PricingAudio Pricing
Chat Speed & LatencyProvider Reliability

Benchmarks

GPQAMMLU-ProAIME 2025SWE-Bench VerifiedMMLUHumanity's Last ExamLiveCodeBenchMATHHumanEvalMMMU-Pro
All Benchmarks

Tools

Model DirectoryCompare Models

Scoring & Data

Scoring & Data
393 models668 benchmarks

Leaderboard Center

Overall RankingCodingCore BenchmarksPrice & ValueRuntime Performance

Modalities

All ModelsImage GenerationImage EditingVideo GenerationImage-to-VideoVideo EditingText-to-SpeechSpeech-to-TextEmbeddings

Data & Methods

Scoring MethodAll BenchmarksReasoningMath

Vendors

All VendorsOpenAIAnthropicGoogle
llmboard.aiCopyright 2026 llmboard.ai

multimodal benchmark

MMMU (val) Leaderboard

Validation set of the Massive Multi-discipline Multimodal Understanding and Reasoning benchmark. Features college-level multimodal questions across 6 core disciplines (Art & Design, Business, Science, Health & Medicine, Humanities & Social Science, Tech & Engineering) spanning 30 subjects and 183 subfields with diverse image types including charts, diagrams, maps, and tables.

Updated Aug 17, 2026

Models13
Model coverage13
MetricScore
EvidenceB

On this page

  • Ranking
  • Highlights
  • Distribution
  • Top models
  • About
  • FAQ

MMMU (val) Ranking

Higher score ranks better on this benchmark.

13 rows
Columns

Show columns

Sort by
Rank
Model
Score
Percentile
Participants
Evidence
Evaluated
Rank01ModelACQwen3 VL 32B ThinkingAlibaba Cloud / Qwen TeamScore78.1%Percentile100.0%Participants13EvidenceCEvaluatedAug 17, 2026
Rank02ModelACQwen3 VL 30B A3B ThinkingAlibaba Cloud / Qwen TeamScore76.0%Percentile91.7%Participants13EvidenceCEvaluatedAug 17, 2026
Rank03ModelACQwen3 VL 32B InstructAlibaba Cloud / Qwen TeamScore76.0%Percentile83.3%Participants13EvidenceCEvaluatedAug 17, 2026
Rank04ModelACQwen3 VL 30B A3B InstructAlibaba Cloud / Qwen TeamScore74.2%Percentile75.0%Participants13EvidenceCEvaluatedAug 17, 2026
Rank05ModelACQwen3 VL 8B ThinkingAlibaba Cloud / Qwen TeamScore74.1%Percentile66.7%Participants13EvidenceCEvaluatedAug 17, 2026
Rank06ModelACQwen3 VL 4B ThinkingAlibaba Cloud / Qwen TeamScore70.8%Percentile58.3%Participants13EvidenceCEvaluatedAug 17, 2026
Rank07ModelACQwen3 VL 8B InstructAlibaba Cloud / Qwen TeamScore69.6%Percentile50.0%Participants13EvidenceCEvaluatedAug 17, 2026
Rank08ModelACQwen3 VL 4B InstructAlibaba Cloud / Qwen TeamScore67.4%Percentile41.7%Participants13EvidenceCEvaluatedAug 17, 2026
Rank09ModelGOGemma 3 27BGoogleScore64.9%Percentile33.3%Participants13EvidenceCEvaluatedAug 17, 2026
Rank10ModelGOGemma 3 12BGoogleScore59.6%Percentile25.0%Participants13EvidenceCEvaluatedAug 17, 2026
Rank11ModelGOGemma 3 4BGoogleScore48.8%Percentile16.7%Participants13EvidenceCEvaluatedAug 17, 2026
Rank12ModelLALFM2.5-VL-3BLiquid AIScore48.4%Percentile8.3%Participants13EvidenceCEvaluatedAug 17, 2026
Rank13ModelCONorth Micro Vision InstructCohereScore32.9%Percentile0.0%Participants13EvidenceCEvaluatedAug 17, 2026

MMMU (val) Highlights

The leading models and scores on this benchmark.

Rank #1Qwen3 VL 32B Thinking78.1%Rank #2Qwen3 VL 30B A3B Thinking76.0%Rank #3Qwen3 VL 32B Instruct76.0%Rank #4Qwen3 VL 30B A3B Instruct74.2%

MMMU (val) Score Distribution

A closer view of the leading scores on this benchmark.

MMMU (val)

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.

Ranking basisThis mmmu (val) 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.

  1. 01
    AC
    Qwen3 VL 32B ThinkingAlibaba Cloud / Qwen Team
    Score
    78.1%

    Strengths

    • Ranks #1 of 13 compared models
    • 100th percentile on this benchmark
    • C evidence result

    Considerations

    • This result measures MMMU (val), not total model capability
  2. 02
    AC
    Qwen3 VL 30B A3B ThinkingAlibaba Cloud / Qwen Team
    Score
    76.0%

    Strengths

    • Ranks #2 of 13 compared models
    • 92th percentile on this benchmark
    • C evidence result

    Considerations

    • This result measures MMMU (val), not total model capability
  3. 03
    AC
    Qwen3 VL 32B InstructAlibaba Cloud / Qwen Team
    Score
    76.0%

    Strengths

    • Ranks #3 of 13 compared models
    • 83th percentile on this benchmark
    • C evidence result

    Considerations

    • This result measures MMMU (val), not total model capability
  4. 04
    AC
    Qwen3 VL 30B A3B InstructAlibaba Cloud / Qwen Team
    Score
    74.2%

    Strengths

    • Ranks #4 of 13 compared models
    • 75th percentile on this benchmark
    • C evidence result

    Considerations

    • This result measures MMMU (val), not total model capability
  5. 05
    AC
    Qwen3 VL 8B ThinkingAlibaba Cloud / Qwen Team
    Score
    74.1%

    Strengths

    • Ranks #5 of 13 compared models
    • 67th percentile on this benchmark
    • C evidence result

    Considerations

    • This result measures MMMU (val), not total model capability

Selection summary

Best AI Models for MMMU (val)

Qwen3 VL 32B Thinking currently leads MMMU (val) with 78.1%. 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.

Benchmark rank #1Qwen3 VL 32B Thinking78.1%Benchmark rank #2Qwen3 VL 30B A3B Thinking76.0%Benchmark rank #3Qwen3 VL 32B Instruct76.0%

What is MMMU (val)?

What MMMU (val) measures and how its scores work.

Validation set of the Massive Multi-discipline Multimodal Understanding and Reasoning benchmark. Features college-level multimodal questions across 6 core disciplines (Art & Design, Business, Science, Health & Medicine, Humanities & Social Science, Tech & Engineering) spanning 30 subjects and 183 subfields with diverse image types including charts, diagrams, maps, and tables.

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
Yes
Evaluation key
mmmu-(val)|llm-stats-current

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 78.1%.

What does MMMU (val) measure?

Validation set of the Massive Multi-discipline Multimodal Understanding and Reasoning benchmark. Features college-level multimodal questions across 6 core disciplines (Art & Design, Business, Science, Health & Medicine, Humanities & Social Science, Tech & Engineering) spanning 30 subjects and 183 subfields with diverse image types including charts, diagrams, maps, and tables.

Is a higher score better?

Yes. Higher values rank better for this benchmark.

How many models are compared?

13 model results are currently shown.

Does this benchmark affect the overall score?

Yes. This benchmark can contribute to the current LLMBoard capability score.