Benchmark
MMAU-Pro Open-ended
Scores are percentages.
- Models measured
- 9
- Scores recorded
- 9
- Sources
- 1
- IQ weight
- 1.72
Score distribution
- 56.5%
- Mean
- 66.8%
- Median
- 17.1%
- Std dev
- 28.5%
- 10th percentile
- 68%
- 90th percentile
- 23.2%
- Min
This benchmark has a weight of 1.72 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
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro Preview· MMAU-Pro data | 68.3% |
| 2 | Gemini 3.5 Flash· MMAU-Pro data | 68% |
| 3 | Gemini 2.5 Flash· MMAU-Pro data | 67.5% |
| 4 | Gemini 3.6 Flash· MMAU-Pro data | 67.4% |
| 5 | Gemini 2.0 Flash· MMAU-Pro data | 66.8% |
| 6 | Gemini 3.7 Flash· MMAU-Pro data | 66.7% |
| 7 | Qwen2.5-Omni 7B· MMAU-Pro data | 52.3% |
| 8 | Gemma 3n E4B IT· MMAU-Pro data | 28.5% |
| 9 | Gemma 3n E2b It· MMAU-Pro data | 23.2% |