AI 评测与排行榜，分别告诉你什么
分清评测基准、一次评测运行与排行榜，用它们回答各自适合的问题。
Model reports
LiveCodeBench
Time-sliced problem set. A score without its problem window is not comparable to any other score.
评测信息与证据
- 来源
- Model reports
- 发布者
- 未收录
- 评测发布日期
- 2024-03-12
- 采集日期
- 未收录
- 模态
- 未收录
本站保留来源报告的结果，不代表独立复测。请先核对任务版本、评测协议、分数单位和报告日期；不同运行条件不能直接合并为统一排名。
原始资料
报告成绩 12
按来源中的观察顺序展示，顺序不代表排名。日期保留原始精度。
| 模型 | 数值 | 单位 | 协议 | 报告日期与口径 | 来源 |
|---|---|---|---|---|---|
| DeepSeek-R1 | 65.9 | percent | Pass@1-COT | 2025-01-22 · document_publication | DeepSeek-R1 |
| DeepSeek-V3 | 37.6 | percent | Pass@1 | 2024-12-27 · document_publication | DeepSeek-V3 |
| DeepSeek-V3 | 40.5 | percent | Pass@1-COT | 2024-12-27 · document_publication | DeepSeek-V3 |
| DeepSeek-V4-Pro | 56.8 | percent | non-think, pass@1 | 2026-04-22 · document_publication | DeepSeek-V4-Pro |
| DeepSeek-V4-Pro | 89.8 | percent | think high, pass@1 | 2026-04-22 · document_publication | DeepSeek-V4-Pro |
| DeepSeek-V4-Pro | 93.5 | percent | think max, pass@1 | 2026-04-22 · document_publication | DeepSeek-V4-Pro |
| DeepSeek-V4-Flash | 91.6 | percent | think max, pass@1 | 2026-06-25 · document_publication | DeepSeek-V4 |
| Gemini 2.5 Pro | 74.2 | percent | 1/1/2025-5/1/2025 window | 2025-06-17 · document_publication | Gemini 2.5 Pro |
| Gemma 4 (31B) | 80.0 | percent | as reported | 2026-03-11 · document_publication | Gemma 4 (31B) |
| Qwen2.5-Coder-32B-Instruct | 31.4 | percent | Pass@1 | 2024-09-18 · document_publication | Qwen2.5-Coder |
| Qwen3.5-397B-A17B | 83.6 | percent | thinking (temp 0.6, top_p 0.95, top_k 20) | 2026-02-16 · document_publication | Qwen3.5-397B-A17B |
| Qwen3-235B-A22B (Thinking) | 70.7 | percent | v5, 2024.10-2025.02 window | 2025-05-14 · document_publication | Qwen3 |