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: 测试正确 ↑.
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） |
|---|---|---|---|---|---|---|---|
| #375 | Gemini 3.1 Flash Lite high | 3 | 0.5 | $2.044 | 10/18 | 62.0s | |
| #84 | GPT-5 Mini medium | OpenAI | 3 | 0.8 | $0.315 | 13/23 | 29.4s |
| #88 | Muse Spark 1.1 high | Meta | 2 | 0.8 | $2.570 | 13/23 | 37.4s |
| #112 | GPT-5.4 Nano medium | OpenAI | 2 | 0.8 | $0.235 | 13/23 | 16.0s |
| #119 | GPT-5.4 Mini medium | OpenAI | 3 | 0.8 | $1.102 | 13/23 | 29.9s |
| #126 | DeepSeek V4.1 Flash 1017 high | DeepSeek | 2 | 0.8 | $0.589 | 13/23 | 58.6s |
| #129 | Step 3.7 Flash medium | Stepfun | 2 | 0.8 | $0.571 | 13/23 | 26.6s |
| #131 | Qwen3.7 Plus none | Qwen | 1 | 0.8 | $0.178 | 13/23 | 17.8s |
| #138 | Claude Haiku 5.5 medium | Anthropic | 2 | 0.8 | $0.063 | 13/23 | 7.71s |
| #152 | Qwen3.6 Flash medium | Qwen | 1 | 0.7 | $0.783 | 13/23 | 45.5s |
| #153 | DeepSeek V4 Flash 0731 low | DeepSeek | 1 | 0.7 | $0.306 | 13/23 | 88.1s |
| #155 | GLM 5.3 FlashX high | Z.ai | 1 | 0.7 | $0.210 | 13/23 | 14.9s |
| #161 | GLM 5.3 Flash high | Z.ai | 2 | 0.7 | $0.056 | 13/23 | 29.1s |
| #173 | Qwen3.7 Flash high | Qwen | 1 | 0.7 | $0.063 | 13/23 | 45.5s |
| #184 | GPT-5.2 Chat default | OpenAI | 1 | 0.7 | $0.594 | 13/23 | 7.43s |