通用智能：未遵循指令
通用智能
未遵循指令
看看哪些 AI 模型在 通用智能 上最容易遇到 未遵循指令，更快找出薄弱点。
78/78
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） |
|---|---|---|---|---|---|---|---|
| #104 | Gemini 3.1 Flash Lite Preview low | 1 | 4.0 | $0.646 | 0/1 | 1.54s | |
| #106 | Gemini 3.1 Flash Lite Preview none | 1 | 4.0 | $0.052 | 0/1 | 741ms | |
| #108 | Ring-2.6-1T medium | Inclusionai | 1 | 4.1 | $0.103 | 0/1 | 58.3s |
| #112 | Claude Sonnet 5 none | Anthropic | 1 | 4.7 | $0.548 | 0/1 | 2.81s |
| #113 | MiMo-V2-Flash medium | Xiaomi | 1 | 4.0 | $0.043 | 0/1 | 4.20s |
| #114 | Qwen3.5-Flash medium | Qwen | 1 | 6.1 | $0.139 | 0/1 | 40.1s |
| #120 | Gemini 3.1 Flash Lite minimal | 1 | 4.0 | $0.047 | 0/1 | 791ms | |
| #121 | gpt-oss-120b medium | OpenAI | 1 | 4.3 | $0.019 | 0/1 | 7.90s |
| #126 | Qwen3.5 Plus 2026-04-20 none | Qwen | 1 | 4.8 | $0.122 | 0/1 | 1.41s |
| #127 | Qwen3.5-35B-A3B none | Qwen | 1 | 6.5 | $0.106 | 0/1 | 1.19s |
| #128 | GPT-5 Nano medium | OpenAI | 1 | 4.1 | $0.114 | 0/1 | 17.5s |
| #130 | Step 3.5 Flash medium | Stepfun | 1 | 5.5 | $0.108 | 0/1 | 22.4s |
| #136 | GPT-5.4 Mini none | OpenAI | 1 | 4.8 | $0.095 | 0/1 | 1.82s |
| #138 | Kimi K2.6 none | Moonshot AI | 1 | 5.4 | $0.184 | 0/1 | 1.55s |
| #140 | Nemotron 3 Super medium | NVIDIA | 1 | 4.1 | $0.050 | 0/1 | 6.91s |