Code Reasoning on CRUX-O
40.75AccuracyHSA-UL
Evaluation results
| Method | Links | |
|---|---|---|
| HSA-ULTraining Strategy=Annealing, Architecture=MoE, Total Params=8B, Activated Params=1B, Training Tokens=8T2025.11 | 40.75 | |
| HSA-ULTraining Strategy=Base, Architecture=MoE, Total Params=8B, Activated Params=1B, Training Tokens=8T2025.11 | 36.31 | |
| TRM-MoETraining Strategy=Base, Architecture=MoE, Total Params=8B, Activated Params=1B, Training Tokens=8T2025.11 | 35.12 | |
| Qwen3Training Strategy=Annealing, Architecture=Dense, Total Params=0.6B, Activated Params=0.6B, Training Tokens=36T2025.11 | 31.62 | |
| HSA-ULTraining Strategy=Annealing, Architecture=Dense, Total Params=0.5B, Activated Params=0.5B, Training Tokens=4T2025.11 | 22.56 | |
| Qwen2.5Training Strategy=Annealing, Architecture=Dense, Total Params=0.5B, Activated Params=0.5B, Training Tokens=18T2025.11 | 14.38 |