MMMU
Harbor: pendingMMMU: 11.5K大学级图文题，30学科
- Samples
- 11,550
- Released
- 2023
- Scores
- 0
How to run
ltb · mmmu-pro
$ npm i -g lettokenburn-cli$ ltb bench pull mmmu-pro
External harness. pip install opencompass && opencompass --models <model_cfg> --datasets MMMU_Pro_10c_vlmevalkit_gen (OpenCompass dataset config dir 'MMMU_Pro') https://github.com/open-compass/opencompass
Runs happen on your machine with your own subscription; only the score summary is pushed. CLI docs
Scores
5 sourced entries| Harness | Source | Confidence | Link | |||
|---|---|---|---|---|---|---|
| GPT-5.4 | 81.2 | N/A | Self-reported | low | Jun 1, 2026 | artificialanalysis.ai |
| Gemini 3.1 Pro | 81.0 | N/A | Self-reported | low | Jun 1, 2026 | artificialanalysis.ai |
| Qwen 3.5 | 79.0 | N/A | Self-reported | low | Jun 1, 2026 | artificialanalysis.ai |
| Claude Opus 4.6 | 77.3 | N/A | Self-reported | low | Jun 1, 2026 | artificialanalysis.ai |
| Claude Sonnet 4.6 | 75.6 | N/A | Self-reported | low | Jun 1, 2026 | artificialanalysis.ai |
Every number links to where it was published. Low-confidence rows are muted and need per-cell verification.
Reproduced by the community
All runsLoading community runs…
About
MMMU-Pro (MMMU-Professional) — 增强版MMMU，去捷径. Scoring: 准确率. Difficulty: 极高. Imported from benchmark_research.md (2026-06).