谜题求解：未遵循指令
谜题求解
未遵循指令
看看哪些 AI 模型在 谜题求解 上最容易遇到 未遵循指令，更快找出薄弱点。
86/86
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 | 分类得分 | 总成本 | 测试正确 | 响应时间（平均） |
|---|---|---|---|---|---|---|---|
| #103 | Qwen3.5-27B none | Qwen | 1 | 6.7 | $0.090 | 1/3 | 1.38s |
| #108 | Ring-2.6-1T medium | Inclusionai | 1 | 5.9 | $0.103 | 1/3 | 20.7s |
| #109 | Mimo V2 PRO medium | Xiaomi | 1 | 6.4 | $0.333 | 1/3 | 5.08s |
| #115 | Gemma 4 31B none | 1 | 6.5 | $0.035 | 1/3 | 4.23s | |
| #121 | gpt-oss-120b medium | OpenAI | 1 | 5.3 | $0.019 | 1/3 | 21.7s |
| #122 | Gemini 3.1 Flash Lite none | 1 | 6.3 | $0.046 | 1/3 | 720ms | |
| #123 | Inkling low | Thinkingmachines | 1 | 6.4 | $0.187 | 1/3 | 2.97s |
| #124 | Qwen3.6 Flash none | Qwen | 1 | 3.5 | $0.062 | 0/3 | 1.21s |
| #127 | Qwen3.5-35B-A3B none | Qwen | 1 | 3.7 | $0.106 | 0/3 | 1.35s |
| #128 | GPT-5 Nano medium | OpenAI | 1 | 5.3 | $0.114 | 1/3 | 20.6s |
| #129 | Nemotron 3 Ultra none | NVIDIA | 1 | 5.9 | $0.095 | 1/3 | 1.06s |
| #130 | Step 3.5 Flash medium | Stepfun | 1 | 5.3 | $0.108 | 1/3 | 7.22s |
| #134 | Mimo V2 Omni medium | Xiaomi | 1 | 5.9 | $0.683 | 1/3 | 2.38s |
| #136 | GPT-5.4 Mini none | OpenAI | 1 | 5.4 | $0.095 | 1/3 | 836ms |
| #138 | Kimi K2.6 none | Moonshot AI | 1 | 3.1 | $0.184 | 0/3 | 1.40s |