Scientific Reasoning on RxnBench Chinese
83.61ScoreQwen3-VL
Evaluation Results
| Method | Links | |
|---|---|---|
| Qwen3-VLparameters=8B2026.01 | 83.61 | |
| MiniCPM-Vversion=4.5, parameters=8B2026.01 | 83.34 | |
| Innovator-VLvariant=8B-Instruct, parameters=8B2026.01 | 82.75 | |
| Intern-S1variant=mini, parameters=9B2026.01 | 82.16 | |
| LLaVA-OVversion=1.5, parameters=8B2026.01 | 81.7 | |
| Innovator-VLvariant=8B-Thinking, parameters=8B2026.01 | 81.11 | |
| MiMo-VLtraining=7B-RL, parameters=7B2026.01 | 80.59 | |
| InternVL3.5parameters=8B2026.01 | 80.26 | |
| MiMo-VLtraining=7B-SFT, parameters=7B2026.01 | 80.26 |