Skip to content

Benchmark

VQAv2

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

Models measured
3
Scores recorded
3
Sources
2
IQ weight
No weight published

Score distribution

82.1%
Mean
80.9%
Median
3.4%
Std dev
78.6%
10th percentile
86.7%
90th percentile
78.6%
Min
78.6%86.7%
The band covers the middle 80% of scores. The line marks the median.

Leaderboard

#ModelScore
1Molmo 72B· EvalAI data86.7%
2Pixtral Large· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim80.9%
3Pixtral 12B2409· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim78.6%

Data credits: Data by LLM Stats