Code Generation on LiveCodeBench v5 2024.10–2025.02
60.6Pass@1DeepCoder-14B
Evaluation Results
| Method | Links | |
|---|---|---|
| DeepCoder-14B2026.03 | 60.6 | |
| EVOLVECODER-4B (r3)Refinement round=r3, Backbone=Qwen3-4B (Thinking)2026.03 | 59.7 | |
| EVOLVECODER-4B (r2)Refinement round=r2, Backbone=Qwen3-4B (Thinking)2026.03 | 59.6 | |
| o12026.03 | 59.5 | |
| Critique-Coder-4B2026.03 | 59 | |
| EVOLVECODER-4B (r1)Refinement round=r1, Backbone=Qwen3-4B (Thinking)2026.03 | 57.2 | |
| EVOLVECODER-4B (r0)Refinement round=r0, Backbone=Qwen3-4B (Thinking)2026.03 | 56 | |
| Qwen3-4B (Thinking)Thinking-mode evaluation prompt=true2026.03 | 54.2 | |
| DeepSeek-R1-Distill-14B2026.03 | 53 | |
| DeepSeek-V2.5-238B2026.03 | 42.6 |