MultimodalEst. 2023

MMMU

Massive Multi-discipline Multimodal Understanding (MMMU) evaluates multimodal models on college-level subject knowledge and deliberate reasoning across 30 subjects and 183 subfields, using images, charts, diagrams, and domain-specific visualizations.

Metrics

Accuracy (%) across 30 college subjects

Created By

IN2 Lab / Waterloo

Top Model Scores

RankModelScoreDate
1GPT-5.274.6%2026-03
2Gemini 3 Ultra73.8%2026-01
3Claude Opus 4.672.1%2026-02
4Grok 468.5%2026-02
5InternVL 365.2%2026-01