Code Verification on LiveCodeBench
39.31Pass@1CVeDRL
Evaluation Results
| Method | Links | |
|---|---|---|
| CVeDRLScale=0.6B, Policy Model=GPT-4o-m2026.01 | 39.31 | |
| CodeRMScale=8.0B, Policy Model=GPT-4o-m2026.01 | 39.18 | |
| Llama3.1Scale=70B, Policy Model=GPT-4o-m2026.01 | 38.61 | |
| Qwen3Scale=32B, Policy Model=GPT-4o-m2026.01 | 38.51 | |
| Qwen3Scale=0.6B, Policy Model=GPT-4o-m2026.01 | 37.18 | |
| CodeTScale=-, Policy Model=GPT-4o-m2026.01 | 35.11 | |
| VanillaScale=-, Policy Model=GPT-4o-m2026.01 | 34.9 | |
| MBR-EScale=-, Policy Model=GPT-4o-m2026.01 | 34.79 | |
| Llama3.1Scale=70B, Policy Model=Llama3 70B2026.01 | 28.39 | |
| CVeDRLScale=0.6B, Policy Model=Llama3 70B2026.01 | 27.96 | |
| Qwen3Scale=32B, Policy Model=Llama3 70B2026.01 | 27.95 | |
| CodeRMScale=8.0B, Policy Model=Llama3 70B2026.01 | 27.81 | |
| CodeTScale=-, Policy Model=Llama3 70B2026.01 | 25.92 | |
| Qwen3Scale=0.6B, Policy Model=Llama3 70B2026.01 | 25.47 | |
| VanillaScale=-, Policy Model=Llama3 70B2026.01 | 25.19 | |
| MBR-EScale=-, Policy Model=Llama3 70B2026.01 | 25.18 | |
| CVeDRLScale=0.6B, Policy Model=GPT-3.52026.01 | 23.09 | |
| Qwen3Scale=32B, Policy Model=GPT-3.52026.01 | 22.86 | |
| Llama3.1Scale=70B, Policy Model=GPT-3.52026.01 | 22.79 | |
| Qwen3Scale=0.6B, Policy Model=GPT-3.52026.01 | 21.91 | |
| CodeRMScale=8.0B, Policy Model=GPT-3.52026.01 | 21.81 | |
| CodeTScale=-, Policy Model=GPT-3.52026.01 | 20.6 | |
| MBR-EScale=-, Policy Model=GPT-3.52026.01 | 20.51 | |
| VanillaScale=-, Policy Model=GPT-3.52026.01 | 20.48 | |
| CVeDRLScale=0.6B, Policy Model=Llama3 8B2026.01 | 16.75 | |
| CodeRMScale=8.0B, Policy Model=Llama3 8B2026.01 | 15.24 | |
| Qwen3Scale=32B, Policy Model=Llama3 8B2026.01 | 15.17 | |
| Llama3.1Scale=70B, Policy Model=Llama3 8B2026.01 | 13.33 | |
| CodeTScale=-, Policy Model=Llama3 8B2026.01 | 12.59 | |
| Qwen3Scale=0.6B, Policy Model=Llama3 8B2026.01 | 12.52 | |
| MBR-EScale=-, Policy Model=Llama3 8B2026.01 | 12.12 | |
| VanillaScale=-, Policy Model=Llama3 8B2026.01 | 11.74 |