Mathematical Reasoning on OpenR1 Math 220k
0.8633Composite ScoreSFTKey-Tag
Evaluation Results
| Method | Links | |
|---|---|---|
| SFTKey-TagModel=Qwen3-8B-Base2025.12 | 0.8633 | |
| SFTKey-TagModel=Qwen2.5-7B2025.12 | 0.7217 | |
| SFT-TagModel=Qwen3-8B-Base2025.12 | 0.6959 | |
| SFT(baseline)Model=Qwen3-8B-Base2025.12 | 0.6164 | |
| Key-TagModel=Qwen3-8B-Base2025.12 | 0.5831 | |
| SFT(baseline)Model=Qwen2.5-7B2025.12 | 0.5772 | |
| SFT-TagModel=SmolLM3-3B-Base2025.12 | 0.5645 | |
| SFT-TagModel=Qwen2.5-7B2025.12 | 0.5604 | |
| SFTKey-TagModel=SmolLM3-3B-Base2025.12 | 0.5225 | |
| SFT(baseline)Model=SmolLM3-3B-Base2025.12 | 0.4574 | |
| Key-TagModel=Qwen2.5-7B2025.12 | 0.3822 | |
| Key-TagModel=Qwen2.5-3B2025.12 | 0.357 | |
| SFTKey-TagModel=Qwen2.5-1.5B2025.12 | 0.3434 | |
| SFTKey-TagModel=Qwen2.5-3B2025.12 | 0.3416 | |
| SFT-TagModel=Qwen2.5-3B2025.12 | 0.3402 | |
| SFT(baseline)Model=Qwen2.5-3B2025.12 | 0.3094 | |
| SFT(baseline)Model=Qwen2.5-1.5B2025.12 | 0.2842 | |
| Key-TagModel=Qwen2.5-1.5B2025.12 | 0.2828 | |
| SFT-TagModel=Qwen2.5-1.5B2025.12 | 0.28 | |
| Key-TagModel=SmolLM3-3B-Base2025.12 | 0.266 |