未遵循指令 失败
看看哪些 AI 模型最常遇到 未遵循指令，让你在选择前先发现稳定性风险。 排序方式: 总成本 ↑.
293/293
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 | 未遵循指令 次数 | 分数 | 总成本 | 测试正确 | 响应时间（平均） |
|---|---|---|---|---|---|---|---|
| #198 | GPT-5.6 Luna low | OpenAI | 1 | 0.7 | $0.049 | 11/23 | 6.96s |
| #330 | Hy3 preview medium | Tencent | 1 | 0.5 | $0.049 | 14/21 | 16.3s |
| #376 | GLM 5V Turbo none | Z.ai | 2 | 0.5 | $0.052 | 8/21 | 2.99s |
| #270 | DeepSeek V4 Flash 0731 none | DeepSeek | 2 | 0.6 | $0.053 | 5/23 | 27.4s |
| #356 | Inkling Small none | Thinkingmachines | 1 | 0.5 | $0.054 | 5/23 | 1.45s |
| #296 | DeepSeek V4.1 Flash 1017 none | DeepSeek | 1 | 0.6 | $0.055 | 8/23 | 3.46s |
| #337 | Inkling Small low | Thinkingmachines | 2 | 0.5 | $0.055 | 9/23 | 1.99s |
| #161 | GLM 5.3 Flash high | Z.ai | 2 | 0.7 | $0.056 | 13/23 | 29.1s |
| #327 | MiMo-V2.6-Flash none | Xiaomi | 1 | 0.5 | $0.060 | 6/23 | 6.20s |
| #75 | Claude Haiku 5.5 low | Anthropic | 2 | 0.9 | $0.061 | 16/23 | 7.51s |
| #173 | Qwen3.7 Flash high | Qwen | 1 | 0.7 | $0.063 | 13/23 | 45.5s |
| #138 | Claude Haiku 5.5 medium | Anthropic | 2 | 0.8 | $0.063 | 13/23 | 7.71s |
| #357 | Qwen3 Coder Next default | Qwen | 1 | 0.5 | $0.067 | 5/23 | 19.7s |
| #207 | Solar Pro 4 xhigh | Upstage | 3 | 0.7 | $0.067 | 12/23 | 142.8s |
| #374 | GPT-5.4 Nano none | OpenAI | 2 | 0.5 | $0.068 | 4/23 | 4.42s |