LiveCodeBench livecodebench Leaderboard
Auto-discovered from papers reporting LiveCodeBench (pass@1). · Metric: pass@1 (higher is better)
| # | Model | pass@1 | Paper |
|---|---|---|---|
| 1 | BenchEvolver: Frontier Task Synthesis via Solution-Centric Evolution | 99.00 | — |
| 2 | CWM: An Open-Weights LLM for Research on Code Generation with World Models | 68.60 | — |
| 3 | CWM: An Open-Weights LLM for Research on Code Generation with World Models | 68.60 | — |
| 4 | Planning In Natural Language Improves LLM Search For Code Generation | 41.40 | — |
| 5 | $V_1$: Unifying Generation and Self-Verification for Parallel Reasoners | 10.00 | — |