Science Question Answering on S-GPQA (accuracy)
57.3Accuracy (S-GPQA)Kimi-K2.5
Evaluation Results
| Method | Links | |
|---|---|---|
| Kimi-K2.5Base Model=Kimi-K2.52026.05 | 57.3 | |
| Qwen3.5-397BBase Model=Qwen3.5-397B2026.05 | 55.4 | |
| Deepseek-V3.2Base Model=Deepseek-V3.22026.05 | 54.6 | |
| Qwen3.5-122BBase Model=Qwen3.5-122B2026.05 | 53 | |
| Qwen3.5-35BBase Model=Qwen3.5-35B2026.05 | 49.7 | |
| GLM-5.1Base Model=GLM-5.12026.05 | 48.9 | |
| GPT-OSS-120BBase Model=GPT-OSS-120B2026.05 | 46.9 | |
| Qwen3.5-9B + MaRBase Model=Qwen3.5-9B, Optimization Method=MaR2026.05 | 46.4 | |
| Qwen3.5-9B + DAPOBase Model=Qwen3.5-9B, Optimization Method=DAPO2026.05 | 44.8 | |
| Qwen3.5-9BBase Model=Qwen3.5-9B2026.05 | 43.3 | |
| Qwen3.5-4B + MaRBase Model=Qwen3.5-4B, Optimization Method=MaR2026.05 | 41.7 | |
| Qwen3.5-4B + DAPOBase Model=Qwen3.5-4B, Optimization Method=DAPO2026.05 | 40.6 | |
| Qwen3.5-4BBase Model=Qwen3.5-4B2026.05 | 38 |