Math Reasoning on MathVision (Accuracy)
72AccuracyGPT5
Evaluation results
| Method | Links | |
|---|---|---|
| GPT5Model Category=Proprietary Model2026.07 | 72 | |
| Gemini-2.5-ProModel Category=Proprietary Model2026.07 | 69.1 | |
| LASERModel Category=Open-source Vision-Language Model2026.07 | 44.6 | |
| Reflection-V-7BModel Category=Open-source Vision-Language Model2026.07 | 33.9 | |
| VAPO-Thinker-7BModel Category=Open-source Vision-Language Model2026.07 | 31.9 | |
| R1-Onevision-7BModel Category=Open-source Vision-Language Model2026.07 | 29.9 | |
| Vision-SR1-7BTraining Data=56K2026.05 | 28.3 | |
| Qwen2.5-VL-7B-Instruct + Faithful-MR1Backbone model=Qwen2.5-VL-7B-Instruct, Training Data=19.2K2026.05 | 28.3 | |
| VisionR1-7BModel Category=Open-source Vision-Language Model2026.07 | 28.2 | |
| Perception-R1-7BTraining Data=1.4K2026.05 | 27.6 | |
| Qwen2.5-VL-7B-Instruct + VPPOBackbone model=Qwen2.5-VL-7B-Instruct, Training Data=19.2K2026.05 | 27.1 | |
| Qwen2.5-VL-7B-Instruct + GRPOBackbone model=Qwen2.5-VL-7B-Instruct, Training Data=19.2K2026.05 | 26.5 | |
| Qwen2.5-VL-7BModel Category=Open-source Vision-Language Model2026.07 | 26.4 | |
| OpenVLThinker-7BModel Category=Open-source Vision-Language Model2026.07 | 25.9 | |
| InternVL2.5-8BModel Category=Open-source Vision-Language Model2026.07 | 25.6 | |
| Qwen2.5-VL-7B-InstructBackbone model=Qwen2.5-VL-7B-Instruct2026.05 | 23.8 | |
| Qwen2.5-VL-3B-Instruct + Faithful-MR1Backbone model=Qwen2.5-VL-3B-Instruct, Training Data=19.2K2026.05 | 23.6 | |
| Qwen2.5-VL-3B-Instruct + VPPOBackbone model=Qwen2.5-VL-3B-Instruct, Training Data=19.2K2026.05 | 21.4 | |
| Qwen2.5-VL-3B-Instruct + GRPOBackbone model=Qwen2.5-VL-3B-Instruct, Training Data=19.2K2026.05 | 20.9 | |
| Qwen2.5-VL-3B-InstructBackbone model=Qwen2.5-VL-3B-Instruct2026.05 | 20.4 |