The AI arena is free today

Open Superagent
AI OrganizationMaintains 7 benchmarks

MMMU Team: AI benchmarks & research

Academic team behind MMMU (Ohio State, CMU, Waterloo and others). Maintains MMMU, MMMU-Pro and Video-MMMU.

Maintained by MMMU Team

7 benchmarks

In one line

MMMU Team maintains MMMU (counts toward the LLM Stats Score, 66 models scored), MMMU-Pro (counts toward the LLM Stats Score, 77 models scored), MMMU-Pro (with tools) (counts toward the LLM Stats Score, 5 models scored), MMMUval (counts toward the LLM Stats Score, 4 models scored), MMMU (val) (counts toward the LLM Stats Score, 14 models scored), MMMU (validation) (counts toward the LLM Stats Score, 4 models scored) and VideoMMMU (counts toward the LLM Stats Score, 29 models scored). It ships no models of its own on LLM Stats.

FAQ

Common questions about MMMU Team.

What is MMMU Team?

Academic team behind MMMU (Ohio State, CMU, Waterloo and others). Maintains MMMU, MMMU-Pro and Video-MMMU. On LLM Stats it maintains 7 benchmarks, led by MMMU-Pro, MMMU and VideoMMMU. It ships no models of its own; each score on its benchmarks carries its own provenance label on the model page.

Which MMMU Team benchmarks count toward the LLM Stats Score?

MMMU-Pro, MMMU, VideoMMMU, MMMU (val), MMMU-Pro (with tools), MMMU (validation) and 1 more count toward the LLM Stats Score. Only benchmarks admitted to the core set move the main leaderboard; the rest are shown for transparency.

Which models lead MMMU Team's benchmarks?

GPT-6.1 Sol leads MMMU-Pro at 86.0% (77 models scored), Qwen3.6 Plus leads MMMU at 86.0% (66 models scored) and Gemini 3 Pro leads VideoMMMU at 87.6% (29 models scored). Each benchmark page lists the full ranking and how the results were produced.

Where are MMMU Team's official benchmark results?

MMMU Team's official site is https://mmmu-benchmark.github.io. Each benchmark page on LLM Stats links to that benchmark's own paper or leaderboard, and shows the scores next to every other benchmark so models can be compared on one page.