Mathematical Reasoning on AIME 2024 (Pass@1 and Token Count)
0.3875Pass@1DeepScaleR-1.5B
Evaluation results
| Method | Links | ||
|---|---|---|---|
| DeepScaleR-1.5BBase Model=DeepScaleR-1.5B2026.02 | 0.3875 | 8,140 | |
| ThinkPrune-Iter2kBase Model=DeepScaleR-1.5B, Fine-tuning Algorithm=ThinkPrune-Iter2k2026.02 | 0.35 | 4,598 | |
| LACONICBase Model=DeepScaleR-1.5B, Fine-tuning Algorithm=LACONIC, Length Budget (B)=20002026.02 | 0.3479 | 3,976 | |
| Efficient-ReasoningBase Model=DeepScaleR-1.5B, Fine-tuning Algorithm=Efficient-Reasoning2026.02 | 0.3271 | 6,440 | |
| DeepSeek-1.5BBase Model=DeepSeek-1.5B2026.02 | 0.3 | 15,218 | |
| GRPOBase Model=DeepScaleR-1.5B, Fine-tuning Algorithm=GRPO2026.02 | 0.2958 | 6,122 | |
| GRPOBase Model=DeepSeek-1.5B, Fine-tuning Algorithm=GRPO2026.02 | 0.2583 | 9,090 | |
| L1-MaxBase Model=DeepScaleR-1.5B, Fine-tuning Algorithm=L1-Max2026.02 | 0.2512 | 2,979 | |
| LACONICBase Model=DeepSeek-1.5B, Fine-tuning Algorithm=LACONIC, Length Budget (B)=15002026.02 | 0.2417 | 5,140 | |
| ThinkPrune-Iter2kBase Model=DeepSeek-1.5B, Fine-tuning Algorithm=ThinkPrune-Iter2k2026.02 | 0.2333 | 5,440 | |
| Efficient-ReasoningBase Model=DeepSeek-1.5B, Fine-tuning Algorithm=Efficient-Reasoning2026.02 | 0.2208 | 8,801 | |
| L1-ExactBase Model=DeepScaleR-1.5B, Fine-tuning Algorithm=L1-Exact2026.02 | 0.2202 | 4,138 |