Mathematical Reasoning on MATH (Pass@1 and Token Usage)
85.92Pass@1DeepScaleR-1.5B
Evaluation results
| Method | Links | ||
|---|---|---|---|
| DeepScaleR-1.5BBase Model=DeepScaleR-1.5B2026.02 | 85.92 | 3,019 | |
| LACONICBase Model=DeepScaleR-1.5B, Fine-tuning Algorithm=LACONIC, Length Budget (B)=20002026.02 | 85.61 | 1,556 | |
| GRPOBase Model=DeepScaleR-1.5B, Fine-tuning Algorithm=GRPO2026.02 | 85.28 | 1,767 | |
| ThinkPrune-Iter2kBase Model=DeepScaleR-1.5B, Fine-tuning Algorithm=ThinkPrune-Iter2k2026.02 | 84.22 | 1,802 | |
| L1-MaxBase Model=DeepScaleR-1.5B, Fine-tuning Algorithm=L1-Max2026.02 | 83.45 | 1,912 | |
| DeepSeek-1.5BBase Model=DeepSeek-1.5B2026.02 | 82.9 | 5,340 | |
| L1-ExactBase Model=DeepScaleR-1.5B, Fine-tuning Algorithm=L1-Exact2026.02 | 82.57 | 3,734 | |
| Efficient-ReasoningBase Model=DeepScaleR-1.5B, Fine-tuning Algorithm=Efficient-Reasoning2026.02 | 82.01 | 1,585 | |
| LACONICBase Model=DeepSeek-1.5B, Fine-tuning Algorithm=LACONIC, Length Budget (B)=15002026.02 | 81.88 | 1,636 | |
| Efficient-ReasoningBase Model=DeepSeek-1.5B, Fine-tuning Algorithm=Efficient-Reasoning2026.02 | 81.63 | 1,909 | |
| GRPOBase Model=DeepSeek-1.5B, Fine-tuning Algorithm=GRPO2026.02 | 81.57 | 2,403 | |
| ThinkPrune-Iter2kBase Model=DeepSeek-1.5B, Fine-tuning Algorithm=ThinkPrune-Iter2k2026.02 | 81.35 | 1,762 |