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
Aggregates ten heterogeneous tool suites; the per-suite spread is wide.
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。
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-V4-Pro | 51.8 | percent | think max, pass@1 | 2026-04-22 · document_publication | DeepSeek-V4-Pro |
| DeepSeek-V4-Pro-0813 | 74.1 | percent | max reasoning effort, temp 1.0, top_p 0.95 | 2026-08-13 · document_publication | DeepSeek-V4-Pro-0813 |
| DeepSeek-V4-Flash | 47.8 | percent | think max, pass@1 | 2026-06-25 · document_publication | DeepSeek-V4 |
| Kimi K3 | 76.5 | percent | reasoning=max | 2026-06-13 · document_publication | Kimi K3 |
| Qwen3.5-397B-A17B | 38.3 | percent | thinking (temp 0.6, top_p 0.95, top_k 20) | 2026-02-16 · document_publication | Qwen3.5-397B-A17B |
| GLM-5.1 | 40.7 | percent | as reported | 2026-04-03 · document_publication | GLM-5.1 |
| GLM-5.2 | 48.2 | percent | official evaluation service, max_token 128K | 2026-06-16 · document_publication | GLM-5.2 |
| GLM-5 | 38.0 | percent | as reported | 2026-02-11 · document_publication | GLM-5 |