Mathematical Reasoning on FlowVerse Text Plus
46.1CoT ErrorInfiMM-Math-7B
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| InfiMM-Math-7BModel Type=Open-source MLLMs2025.03 | 46.1 | 40.1 | |
| InternVL2.5-8BModel Type=Open-source MLLMs2025.03 | 49.6 | 42.7 | |
| Qwen-VL-MaxModel Type=Closed-source MLLMs2025.03 | 53.9 | 51 | |
| GPT-4o-miniModel Type=Closed-source MLLMs2025.03 | 59.6 | 55.2 | |
| Qwen2.5-VL-7BModel Type=Open-source MLLMs2025.03 | 62 | 55 | |
| GPT-4oModel Type=Closed-source MLLMs2025.03 | 62.2 | 58.2 | |
| Qwen2-VL-72BModel Type=Open-source MLLMs2025.03 | 63.7 | 50 | |
| Claude-3.5-SonnetModel Type=Closed-source MLLMs2025.03 | 64 | 58.3 | |
| MathFlowModel Type=Open-source MLLMs, Backbone=Qwen2.5-VL-7B2025.03 | 64.2 | 56 | |
| VLM-R1-7B†Model Type=Open-source MLLMs2025.03 | 65.5 | 58.9 | |
| InternVL2.5-78BModel Type=Open-source MLLMs2025.03 | 67.8 | 64.7 | |
| Gemini-2.5-proModel Type=Closed-source MLLMs2025.03 | 68.9 | 64.1 | |
| GPT-4VModel Type=Closed-source MLLMs2025.03 | 72 | 61.4 | |
| GPT-5Model Type=Closed-source MLLMs2025.03 | 77 | 69.2 | |
| MathFlowModel Type=Closed-source MLLMs, Backbone=GPT-52025.03 | 77 | 69.3 |