MMMU-Pro
Harder MMMU variant
multimodal
Definition
- Version
- not in the notes
- Category
- multimodal
- Maintainer
- MMMU team
- Status
- not in the notes
- Home page
- not in the notes
Sources
- huggingface.co/MiniMaxAI/MiniMax-M3 https://huggingface.co/MiniMaxAI/MiniMax-M3
- MMMU Leaderboard & Scores , October 2026 benchlm.ai https://www.benchlm.ai/benchmarks/mmmu
- OpenBMB launches MiniCPM-V 4.6 1.3B Instruct artificialanalysis.ai https://www.artificialanalysis.ai/articles/openbmb-launches-minicpm-v-4-6-1-3b-instruct
Scores at a Glance
Score in %
Higher is better for this unit.
1
MiniMax-M3MiniMax
MiniCPM-V 4.6 1.3BOpenBMB (Tsinghua NLP Lab and ModelBest)Reported in Artificial Analysis article; run origin not stated
MiniMax-M3MiniMax78.1 %
vendor-reported2MiniCPM-V 4.6 1.3BOpenBMB (Tsinghua NLP Lab and ModelBest)Reported in Artificial Analysis article; run origin not stated38 %
independent0255075100
Where this comes from
Each bar is one model, with its most independent recorded score. Hatched bars are vendor-reported; the word beside each bar says who ran it. Settings and dates differ between rows; the Scores table below has every result.
Scores
| Model | Score | Who ran it | Setting | View or population | Date | Source |
|---|---|---|---|---|---|---|
| MiniMax-M3 | 78.1 % | vendor-reported | – | – | – | huggingface.co |
| MiniCPM-V 4.6 1.3B | 38 % | independent | Reported in Artificial Analysis article; run origin not stated | – | 11 May 2026 | artificialanalysis.ai |
Where this comes from
Scores from different settings, harnesses or dates are not directly comparable; the row says which. Sorted by score, highest first, as a reading aid, not a ranking.
To cite this page: Project AURION, AI Use-Cases Research & Innovation, "MMMU-Pro", https://aurion.semai.id/benchmarks/mmmu-pro/, records of October 2026.