Benchmark
MLVU(M-Avg)
Scores are percentages. Higher is better.
Benchmark sourceScores from BenchBoard docs.
- Models measured
- 8
- Scores recorded
- 8
- Sources
- 1
- IQ weight
- 2.54
Score distribution
- 86.6%
- Mean
- 86.7%
- Median
- 2.3%
- Std dev
- 86.2%
- 10th percentile
- 87.4%
- 90th percentile
- 81.9%
- Min
This benchmark has a weight of 2.54 for chat models, measured over 5 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 | Qwen3.8 MaxQwen3.8-Max release vision/video table · published setting· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 90.8% |
| 2 | Qwen3.7 PlusQwen3.7-Plus release · published setting· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 87.4% |
| 3 | Qwen3.5 Omni PlusQwen3.5-Omni release · published setting· Reported by Qwen, Qwen Team· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 86.8% |
| 4 | Qwen3.5 397B A17BQwen3.5 release/model card · published setting· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 86.7% |
| 5 | Qwen3.6 PlusQwen official release table · thinking mode unless benchmark requires otherwise· Reported by Alibaba Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 86.7% |
| 6 | Qwen3.6 27BQwen official release table · thinking mode unless benchmark requires otherwise· Reported by Alibaba Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 86.6% |
| 7 | Qwen3.6 FlashQwen3.6-35B-A3B official model card · hosted API name Qwen3.6-Flash· Reported by Alibaba Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 86.2% |
| 8 | Qwen3.5 Omni FlashQwen3.5-Omni release · published setting· Reported by Qwen, Qwen Team· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 81.9% |