STEM & Puzzle on DynaMath (test)
83.4AccuracyMMFineReason-8B
Evaluation results
| Method | Links | |
|---|---|---|
| MMFineReason-8BModel Source Category=Ours, Model Parameters=8B, Evaluation Mode=Thinking Mode2026.01 | 83.4 | |
| Qwen3-VL 32BModel Source Category=Open-weight VLMs, Model Parameters=32B, Evaluation Mode=Thinking Mode2026.01 | 82 | |
| GPT5 miniModel Source Category=Closed-source VLMs, Evaluation Mode=Thinking Mode2026.01 | 81.4 | |
| MMFineReason-4BModel Source Category=Ours, Model Parameters=4B, Evaluation Mode=Thinking Mode2026.01 | 80.6 | |
| Qwen3-VL 30B-A3BModel Source Category=Open-weight VLMs, Model Parameters=30B-A3B, Evaluation Mode=Thinking Mode2026.01 | 80.1 | |
| Gemini-2.5 FlashModel Source Category=Closed-source VLMs, Evaluation Mode=Thinking Mode2026.01 | 75.9 | |
| MMR1 8BModel Source Category=Open-source VLMs, Model Parameters=8B, Evaluation Mode=Thinking Mode2026.01 | 73.6 | |
| Qwen3-VL 8BModel Source Category=Open-weight VLMs, Model Parameters=8B, Evaluation Mode=Thinking Mode2026.01 | 73.2 | |
| MMFineReason-2BModel Source Category=Ours, Model Parameters=2B, Evaluation Mode=Thinking Mode2026.01 | 71.4 | |
| HoneyBee 8BModel Source Category=Open-source VLMs, Model Parameters=8B, Evaluation Mode=Thinking Mode2026.01 | 69.4 | |
| OMR 7BModel Source Category=Open-source VLMs, Model Parameters=7B, Evaluation Mode=Thinking Mode2026.01 | 69.1 |