反AI技巧:未遵循指令
反AI技巧
未遵循指令
看看哪些 AI 模型在 反AI技巧 上最容易遇到 未遵循指令,更快找出薄弱点。
49/49
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) |
|---|---|---|---|---|---|---|---|
| #341 | Mercury 2.5 Preview low | Inception | 1 | 0.6 | $0.015 | 1/4 | 759ms |
| #342 | MiniMax M2.7 medium | Minimax | 1 | 0.8 | $0.326 | 2/4 | 40.3s |
| #343 | Dots 3 Note Preview default | Dots Studio | 1 | 0.3 | $0.000 | 0/4 | 1.92s |
| #348 | Qwen3 Coder Next default | Qwen | 1 | 0.4 | $0.067 | 0/4 | 3.31s |
| #363 | Qwen3 Coder Next medium | Qwen | 1 | 0.3 | $0.071 | 0/4 | 8.64s |
| #372 | Owl Alpha default | Openrouter | 1 | 0.3 | $0.000 | 0/4 | 2.78s |
| #373 | Ling 2.6 Flash default | Inclusionai | 1 | 0.7 | $0.002 | 2/4 | 11.8s |
| #375 | Ring 2.6 1t default | Inclusionai | 1 | 0.9 | $0.026 | 3/4 | 43.3s |
| #376 | Mimo V2 Omni default | Xiaomi | 1 | 0.4 | $0.021 | 0/4 | 1.63s |
| #380 | Granite 4.2 8B none | IBM Granite | 1 | 0.5 | $0.037 | 1/4 | 17.8s |
| #384 | MiniMax M2.5 medium | Minimax | 1 | 0.8 | $0.440 | 2/4 | 20.8s |
| #385 | GLM 4.7 Flash medium | Z.ai | 1 | 0.5 | $0.193 | 1/4 | 15.0s |
| #386 | gpt-oss-120b none | OpenAI | 1 | 0.7 | $0.016 | 2/4 | 32.8s |
| #388 | Mercury 2.5 none | Inception | 1 | 0.4 | $0.026 | 0/4 | 469ms |
| #396 | Granite 4.1 8b default | IBM Granite | 1 | 0.5 | $0.007 | 1/4 | 844ms |