Skip to content

Benchmark

BLINK - Relative_Reflectance

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

38.2%
Mean
46.3%
Median
9.6%
Std dev
27.6%
10th percentile
49.3%
90th percentile
27.6%
Min

Leaderboard

#ModelScore
1Qwen-VL MaxRelative_Reflectance· EvalAI data49.3%
2Gemini ProRelative_Reflectance· EvalAI data46.3%
3LLaVA 1.6 34BRelative_Reflectance· EvalAI data29.9%
4Claude 3 OpusRelative_Reflectance· EvalAI data27.6%