指令遵循 排名
看看哪些 AI 模型在 指令遵循 上表现最好，哪些更稳定，以及差距主要出现在哪里。 排序方式: 测试正确 ↓.
216/216
筛选模型
没有模型匹配当前搜索和筛选条件。
| 排名 | 模型 | 公司 | 指令遵循 得分 | 分数 | 总成本 | 测试正确 | 响应时间（平均） |
|---|---|---|---|---|---|---|---|
| #85 | KAT-Coder-Pro V2.5 medium | Kwaipilot | 9.9 | 6.9 | $0.467 | 2/2 | 2.58s |
| #88 | MiMo-V2.5-Pro medium | Xiaomi | 9.9 | 6.9 | $0.187 | 2/2 | 2.77s |
| #89 | Qwen3.6 Flash medium | Qwen | 10.0 | 6.9 | $0.738 | 2/2 | 6.05s |
| #90 | Step 3.7 Flash high | Stepfun | 9.8 | 6.9 | $1.207 | 2/2 | 1.52s |
| #94 | Qwen3.6 35B A3B medium | Qwen | 10.0 | 6.7 | $0.746 | 2/2 | 7.50s |
| #95 | Gemini 3.5 Flash-Lite low | 9.8 | 6.7 | $0.145 | 2/2 | 869ms | |
| #97 | KAT-Coder-Pro V2.5 none | Kwaipilot | 9.8 | 6.7 | $0.476 | 2/2 | 2.61s |
| #98 | GLM 5V Turbo medium | Z.ai | 9.9 | 6.7 | $0.457 | 2/2 | 3.74s |
| #99 | Claude Opus 4.7 none | Anthropic | 10.0 | 6.6 | $0.505 | 2/2 | 1.46s |
| #100 | Gemma 4 26B A4B medium | 10.0 | 6.6 | $0.089 | 2/2 | 17.5s | |
| #101 | GLM 5.2 none | Z.ai | 9.8 | 6.6 | $0.128 | 2/2 | 3.84s |
| #103 | Qwen3.6 Max Preview none | Qwen | 9.8 | 6.6 | $0.231 | 2/2 | 1.40s |
| #104 | Gemini 3.5 Flash-Lite medium | 9.8 | 6.5 | $0.369 | 2/2 | 1.59s | |
| #105 | Qwen3.6 27B medium | Qwen | 10.0 | 6.5 | $0.779 | 2/2 | 38.0s |
| #106 | Hy3 preview medium | Tencent | 10.0 | 6.5 | $0.018 | 2/2 | 6.16s |