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
Per-language resolution rates differ widely, so an average hides which languages the model actually handles.
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 | 69.8 | percent | non-think, resolved | 2026-04-22 · document_publication | DeepSeek-V4-Pro |
| DeepSeek-V4-Pro | 74.1 | percent | think high, resolved | 2026-04-22 · document_publication | DeepSeek-V4-Pro |
| DeepSeek-V4-Pro | 76.2 | percent | think max, resolved | 2026-04-22 · document_publication | DeepSeek-V4-Pro |
| DeepSeek-V4-Flash | 73.3 | percent | think max, resolved | 2026-06-25 · document_publication | DeepSeek-V4 |
| Qwen3.5-397B-A17B | 69.3 | percent | thinking (temp 0.6, top_p 0.95, top_k 20) | 2026-02-16 · document_publication | Qwen3.5-397B-A17B |
| Hy4 preview | 82.9 | percent | as printed in the report's benchmark table | 2026-08-28 · document_publication | Hy4 preview |
| GLM-5 | 73.3 | percent | OpenHands, tailored instruction prompt, temp 0.7, top_p 0.95, 200K context | 2026-02-11 · document_publication | GLM-5 |