mistralai/Pixtral-Large-Instruct-2411
| benchmark | category | score | context | conf | evidence |
|---|---|---|---|---|---|
| DocVQA | multimodal | 92.3 | MathVista (CoT) | MMMU (CoT) | ChartQA (CoT) | low | | Gemini-1.5 Pro (measured) | 67.8 | 66.3 | 83.8 | 92.3 | 70.6 | **<u>94.6</u>** | 6.8 | |
| ChartQA | multimodal | 88.1 | MathVista (CoT) | MMMU (CoT) | ChartQA (CoT) | low | | **Pixtral Large (124B)** | **<u>69.4</u>** | 64.0 | 88.1 | **<u>93.3</u>** | **<u>80.9</u>** | 93.8 | **<u>7.4</u>** | |
| MathVista | multimodal | 67.8 | MathVista (CoT) | MMMU (CoT) | ChartQA (CoT) | low | | Gemini-1.5 Pro (measured) | 67.8 | 66.3 | 83.8 | 92.3 | 70.6 | **<u>94.6</u>** | 6.8 | |
| MMMU | multimodal | 64.0 | MathVista (CoT) | MMMU (CoT) | ChartQA (CoT) | low | | **Pixtral Large (124B)** | **<u>69.4</u>** | 64.0 | 88.1 | **<u>93.3</u>** | **<u>80.9</u>** | 93.8 | **<u>7.4</u>** | |
| MT-Bench | chat | 6.8 | MathVista (CoT) | MMMU (CoT) | ChartQA (CoT) | low | | Gemini-1.5 Pro (measured) | 67.8 | 66.3 | 83.8 | 92.3 | 70.6 | **<u>94.6</u>** | 6.8 | |
none scraped.