Science Question Answering on BioPro
70AccuracyKimi-K2.5
Evaluation Results
| Method | Links | |
|---|---|---|
| Kimi-K2.5Base Model=Kimi-K2.52026.05 | 70 | |
| Qwen3.5-397BBase Model=Qwen3.5-397B2026.05 | 69.4 | |
| Qwen3.5-122BBase Model=Qwen3.5-122B2026.05 | 69.2 | |
| Deepseek-V3.2Base Model=Deepseek-V3.22026.05 | 68.5 | |
| GLM-5.1Base Model=GLM-5.12026.05 | 67.5 | |
| Qwen3.5-35BBase Model=Qwen3.5-35B2026.05 | 67.3 | |
| GPT-OSS-120BBase Model=GPT-OSS-120B2026.05 | 63.5 | |
| Qwen3.5-9B + MaRBase Model=Qwen3.5-9B, Optimization Method=MaR2026.05 | 62.8 | |
| Qwen3.5-9B + DAPOBase Model=Qwen3.5-9B, Optimization Method=DAPO2026.05 | 61.7 | |
| Qwen3.5-9BBase Model=Qwen3.5-9B2026.05 | 61.5 | |
| Qwen3.5-4B + MaRBase Model=Qwen3.5-4B, Optimization Method=MaR2026.05 | 58.8 | |
| Qwen3.5-4B + DAPOBase Model=Qwen3.5-4B, Optimization Method=DAPO2026.05 | 58.5 | |
| Qwen3.5-4BBase Model=Qwen3.5-4B2026.05 | 58 |