Multimodal Mathematical Reasoning on DynaMath-W
60.5AccuracySeed-1.5-thinking
Evaluation Results
| Method | Links | |
|---|---|---|
| Seed-1.5-thinkingAccess Type=Closed-Source2025.12 | 60.5 | |
| Gemini-2.5-Pro-ThinkingAccess Type=Closed-Source2025.12 | 57.1 | |
| GPT-4.1Access Type=Closed-Source2025.12 | 48.3 | |
| Claude-4-SonnetAccess Type=Closed-Source2025.12 | 46.9 | |
| OursAccess Type=Open-Source, Model Scale=7B2025.12 | 30.3 | |
| Revisual-R1Access Type=Open-Source, Model Scale=7B2025.12 | 27.5 | |
| VL-RethinkerAccess Type=Open-Source, Model Scale=7B2025.12 | 27.4 | |
| Vision-G1Access Type=Open-Source, Model Scale=7B2025.12 | 27.2 | |
| MMR1-Math-v0Access Type=Open-Source, Model Scale=7B2025.12 | 25.2 | |
| MM-EurekaAccess Type=Open-Source, Model Scale=7B2025.12 | 24 | |
| VLAA-ThinkingAccess Type=Open-Source, Model Scale=7B2025.12 | 22.4 | |
| ThinkLite-VLAccess Type=Open-Source, Model Scale=7B2025.12 | 22 | |
| Qwen2.5-VLAccess Type=Open-Source, Model Scale=7B2025.12 | 21.2 | |
| OpenVLThinkerAccess Type=Open-Source, Model Scale=7B2025.12 | 21.2 |