Mathematical Reasoning on AIME 2024 (pass@1 avg@64)
90.5Pass@1 (avg@64)Klear-Reasoner-8B
Evaluation Results
| Method | Links | |
|---|---|---|
| Klear-Reasoner-8BMaximum Inference Length=64K, Temperature=0.6, Top-p=0.952025.08 | 90.5 | |
| Deepseek-R1-0528-Distill-8BMaximum Inference Length=64K, Temperature=0.6, Top-p=0.952025.08 | 86 | |
| OpenReasoning-Nemotron-7BMaximum Inference Length=64K, Temperature=0.6, Top-p=0.952025.08 | 84.7 | |
| Klear-Reasoner-8BMaximum Inference Length=32K, Temperature=0.6, Top-p=0.952025.08 | 83.2 | |
| POLARIS-4B-PreviewMaximum Inference Length=96K, Temperature=0.6, Top-p=0.952025.08 | 81.2 | |
| Qwen3-8BMaximum Inference Length=32K, Temperature=0.6, Top-p=0.952025.08 | 76 | |
| Klear-Reasoner-8B-SFTMaximum Inference Length=32K, Temperature=0.6, Top-p=0.952025.08 | 75.6 | |
| AceReason-Nemotron-1.1-7BMaximum Inference Length=32K, Temperature=0.6, Top-p=0.952025.08 | 72.6 | |
| Skywork-OR1-7BMaximum Inference Length=32K, Temperature=0.6, Top-p=0.952025.08 | 70.2 | |
| MiMo-7B-RLMaximum Inference Length=32K, Temperature=0.6, Top-p=0.952025.08 | 68.2 | |
| AReal-boba-RL-7BMaximum Inference Length=32K, Temperature=0.6, Top-p=0.952025.08 | 61.9 |