Code Generation on LiveCodeBench Easy
100Pass@1Gemini-2.5
Evaluation Results
| Method | Links | |
|---|---|---|
| Gemini-2.5Evaluation Protocol=Leaderboard2025.12 | 100 | |
| O3Evaluation Protocol=Leaderboard2025.12 | 100 | |
| O4-mini-highEvaluation Protocol=Leaderboard2025.12 | 100 | |
| ORMEvaluation Protocol=Test-Time Scaling, Base Model=O4-mini-high2025.12 | 100 | |
| DreamPRM-CodeEvaluation Protocol=Test-Time Scaling, Base Model=O4-mini-high2025.12 | 100 | |
| DeepSeek-R1Evaluation Protocol=Leaderboard2025.12 | 99.7 | |
| DEPO-iter3Base Model=Qwen2.5-Coder-7B-Instruct, Iteration=32026.05 | 65.3 | |
| XPO-iter3Base Model=Qwen2.5-Coder-7B-Instruct, Iteration=32026.05 | 58.4 | |
| POPO-iter3Base Model=Qwen2.5-Coder-7B-Instruct, Iteration=32026.05 | 58.2 | |
| DPO-iter3Base Model=Qwen2.5-Coder-7B-Instruct, Iteration=32026.05 | 57.6 | |
| Qwen2.5-Coder-7B-Instruct2026.05 | 56.1 |