Skip to content

Benchmark

InfoVQAtest

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

Models measured
8
Scores recorded
10
Sources
1
IQ weight
1.14

Score distribution

86.9%
Mean
87%
Median
3.1%
Std dev
83.1%
10th percentile
89.5%
90th percentile
82%
Min
82%92.6%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.14 for chat models, measured over 8 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
1Kimi K2.5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim92.6%
2Qwen3 VL 235Bthinking· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it.89.5%
3Qwen3 VL 235B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it.89.2%
4Qwen3 VL 32B Thinking· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it.89.2%
5Qwen3 VL 32B Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it.87%
6Qwen: Qwen3 VL 8B Thinking· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it.86%
7qwen3-vl-30bthinking· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it.86%
8Qwen2-VL-72B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim84.5%
9Qwen3 VL 8B Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it.83.1%
10qwen3-vl-30b· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it.82%

Data credits: Data by LLM Stats