Skip to content

Benchmark

MMMU (validation)

Vision benchmarks measure how well models understand images. Scores are percentages.

Models measured
4
Scores recorded
4
Sources
1
IQ weight
No weight published

Score distribution

76.9%
Mean
77.1%
Median
2.7%
Std dev
73.2%
10th percentile
80.7%
90th percentile
73.2%
Min
73.2%80.7%
The band covers the middle 80% of scores. The line marks the median.

Leaderboard

#ModelScore
1Claude Opus 4.5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim80.7%
2Claude Opus 4.1· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim77.1%
3Claude Opus 4· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim76.5%
4Claude Haiku 4.5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim73.2%

Data credits: Data by LLM Stats