Mathematical Reasoning on AIME 2425
93.54Pass@1Gemini
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| GeminiBackbone=Qwen3-235B2026.02 | 93.54 | 18.5 | |
| Free()LMBackbone=Qwen3-235B2026.02 | 93.13 | 16.4 | |
| VanillaBackbone=Qwen3-235B2026.02 | 92.29 | 21.7 | |
| No TrainBackbone=Qwen3-235B2026.02 | 90.63 | 16.2 | |
| GeminiBackbone=Qwen3-30B2026.02 | 87.92 | 15.4 | |
| Free()LMBackbone=Qwen3-30B2026.02 | 87.92 | 14.5 | |
| No TrainBackbone=Qwen3-30B2026.02 | 85.21 | 15.9 | |
| VanillaBackbone=Qwen3-30B2026.02 | 83.33 | 14.9 | |
| Free()LMBackbone=Qwen3-8B2026.02 | 75 | 13 | |
| No TrainBackbone=Qwen3-8B2026.02 | 73.33 | 15.3 | |
| VanillaBackbone=Qwen3-8B2026.02 | 71.67 | 17.6 | |
| GeminiBackbone=Qwen3-8B2026.02 | 71.67 | 13.7 | |
| H2OBackbone=Qwen3-8B2026.02 | 60 | 17.8 | |
| TCBackbone=Qwen3-8B2026.02 | 51.67 | 20.3 |