Visual Math Reasoning on MathVision
72ScoreGPT-5-Thinking
Evaluation Results
| Method | Links | |
|---|---|---|
| GPT-5-ThinkingModel Category=Close-source Models2025.09 | 72 | |
| Gemini-2.5-ProModel Category=Close-source Models2025.09 | 69.1 | |
| QvQ-72B-PreviewModel Category=Open-source Models2025.10 | 34.9 | |
| VAPO-Thinker-7BModel Category=Our Models2025.09 | 31.9 | |
| R1-OneVision-7BModel Category=Open-source Models2025.09 | 29.9 | |
| R1-Onevision-7BModel Category=Reasoning Fine-tuning Methods2025.10 | 29.9 | |
| X-REASONERModel Category=Reasoning Fine-tuning Methods2025.10 | 29.6 | |
| InternVL3-8BModel Category=Open-source Models2025.10 | 28.6 | |
| Vision-R1-7BModel Category=Open-source Models2025.09 | 28.2 | |
| DRIFTModel Category=Reasoning Fine-tuning Methods2025.10 | 26.6 | |
| VLAA-Thinker-7BModel Category=Open-source Models2025.09 | 26.4 | |
| InternVL2.5-8BModel Category=Open-source Models2025.09 | 25.6 | |
| SFT BaselineModel Category=Reasoning Fine-tuning Methods2025.10 | 25.1 | |
| Qwen2.5-VL-7BModel Category=Open-source Models, Evaluation Source=Reproduced by authors2025.10 | 25 | |
| R1-VL-7BModel Category=Reasoning Fine-tuning Methods2025.10 | 24.7 | |
| VAPO-Thinker-3BModel Category=Our Models2025.09 | 23.9 | |
| Qwen2.5-VL-7BModel Category=Open-source Models2025.09 | 23.2 | |
| OpenVLThinker-7BModel Category=Reasoning Fine-tuning Methods2025.10 | 23 | |
| Kimi-VL-16BModel Category=Open-source Models2025.10 | 21.8 | |
| InternVL2-8BModel Category=Open-source Models2025.10 | 20 | |
| Qwen2-VL-7BModel Category=Open-source Models2025.10 | 19.2 | |
| InternLM-XComposer2.5Model Category=Open-source Models2025.10 | 17.8 | |
| LLaVA-OneVision-7BModel Category=Open-source Models2025.10 | 17.6 | |
| InternVL2.5-8BModel Category=Open-source Models2025.10 | 17 |