LiveCodeBench - August to November 2024
Code
LiveCodeBench code-generation problems in the August 1 to November 1, 2024 date window, which Qwen labels "LiveCodeBench 2024.08-2024.11" and DeepSeek's V3 report labels 0801-1101 or August to November 2024. This exact historical problem cohort is retained as a diagnostic rather than pooled with other LiveCodeBench windows.
Top models (higher is better)
| Model | Score |
|---|---|
| o1-mini | 58.0 |
| o1 Preview | 53.6 |
| QwQ-32B-Preview | 50.0 |
| Qwen2.5-Max | 38.7 |
| GPT-4o (2024-08-06) | 35.1 |
| GPT-4o (2024-05-13) | 33.4 |
| DeepSeek-V2.5 | 29.2 |
| DeepSeek-V2-Chat | 18.8 |