General Multimodal Reasoning on M3CoT
78.21Pass@1 AccuracyMiMo-VL-7B-OpenMMReasoner
Evaluation results
| Method | Links | |
|---|---|---|
| MiMo-VL-7B-OpenMMReasonerSeries=MiMo-VL-7B2026.02 | 78.21 | |
| MiMo-VL-7B-DeepVisionSeries=MiMo-VL-7B, Training Algorithm=GSPO2026.02 | 72.56 | |
| Qwen3-VL-8B-DeepVisionSeries=Qwen3-VL-8B, Training Algorithm=GSPO2026.02 | 71.61 | |
| Qwen3-VL-8B-ThinkingSeries=Qwen3-VL-8B2026.02 | 71.31 | |
| Qwen3-VL-8B-InstructSeries=Qwen3-VL-8B2026.02 | 70.83 | |
| MiMo-VL-7B-RL-2508Series=MiMo-VL-7B2026.02 | 70.57 | |
| MiMo-VL-7B-MM-EurekaSeries=MiMo-VL-7B2026.02 | 70.36 | |
| MiMo-VL-7B-MathBookSeries=MiMo-VL-7B2026.02 | 70.23 | |
| MiMo-VL-7B-SFT-2508Series=MiMo-VL-7B2026.02 | 70.02 | |
| GPT-5-Nano-HighModel Category=Closed-source Models2026.02 | 69.15 | |
| Gemini-2.5-Flash-LiteModel Category=Closed-source Models2026.02 | 68.42 |