Scientific Reasoning on SFE
44.06AccuracyGPT-5
Evaluation Results
| Method | Links | |
|---|---|---|
| GPT-52026.06 | 44.06 | |
| S1-VL-32B-RL2026.06 | 43.1 | |
| S1-VL-32B-SFT2026.06 | 42.58 | |
| S1-Omni-Image2026.06 | 42.32 | |
| Qwen3-VL-235B-A22B-Thinking2026.06 | 39.98 | |
| Qwen3-VL-32B-Thinking2026.06 | 37.5 | |
| Thyme-VL2026.06 | 25.37 |