Multimodal Math Reasoning on V-Math
53AccuracyAutoTool (Qwen3-8B)
Evaluation Results
| Method | Links | |
|---|---|---|
| AutoTool (Qwen3-8B)Size=8B, Backbone=Qwen32025.12 | 53 | |
| AutoTool (Qwen2.5-VL-7B)Size=7B, Backbone=Qwen2.5-VL2025.12 | 44.3 | |
| GPT4oSize=-2025.12 | 41.4 | |
| Qwen2.5-VL-72B-InstructSize=72B2025.12 | 24.5 | |
| v-ToolRLSize=2B2025.12 | 19.5 |