Benchmark
MLVU
Vision benchmarks measure how well models understand images. Scores are percentages.
All vision benchmarksScores from LLM Stats data.
- Models measured
- 9
- Scores recorded
- 10
- Sources
- 1
- IQ weight
- 2.19
Score distribution
- 84.4%
- Mean
- 86.2%
- Median
- 4.9%
- Std dev
- 83.8%
- 10th percentile
- 87.3%
- 90th percentile
- 70.2%
- Min
This benchmark has a weight of 2.19 for chat models, measured over 9 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.
Leaderboard
| # | Model | Score |
|---|---|---|
| 1 | Qwen3.7 Plus· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 87.4% |
| 2 | Qwen3.5 122B-A10B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 87.3% |
| 3 | Qwen3.6 Plus· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 86.7% |
| 4 | Qwen3.6 27B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 86.6% |
| 5 | Qwen3.6 35B A3B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 86.2% |
| 6 | Qwen3.5 27B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 85.9% |
| 7 | Qwen3.5 35B A3B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 85.6% |
| 8 | Qwen3 VL 235B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. | 84.3% |
| 9 | Qwen3 VL 235Bthinking· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. | 83.8% |
| 10 | Qwen2.5-VL 7B Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 70.2% |
Data credits: Data by LLM Stats