Skip to content

Benchmark

Vibe-Eval

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

Models measured
7
Scores recorded
7
Sources
1
IQ weight
1.48

Score distribution

56.2%
Mean
56.3%
Median
8.7%
Std dev
48.9%
10th percentile
65.6%
90th percentile
40.9%
Min
40.9%67.2%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.48 for chat models, measured over 7 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
1Gemini 2.5 Pro· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim67.2%
2Gemini 2.5 Flash· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim65.4%
3Gemini 2.0 Flash· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim56.3%
4Gemini 1.5 Pro· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim53.9%
5Gemini 2.5 Flash-Lite· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim51.3%
6Gemini 1.5 Flash· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim48.9%
7Gemini 1.5 Flash-8B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim40.9%

Data credits: Data by LLM Stats