Did not follow instructions Failures
See which AI models run into Did not follow instructions most often, so you can spot reliability risks before choosing one. Sort by: Total Cost ↓.
141/141
Filter models
No models match the current search and filters.
| Rank | Model | Company | Did not follow instructions Count | Score | Total Cost | Tests Correct | Response Time (avg) |
|---|---|---|---|---|---|---|---|
| #112 | Gemini 3.1 Flash Lite Preview none | 2 | 6.4 | $0.052 | 12/22 | 1.58s | |
| #151 | GLM 5V Turbo none | Z.ai | 2 | 5.6 | $0.052 | 8/21 | 2.99s |
| #155 | KAT-Coder-Air V2.5 medium | Kwaipilot | 1 | 5.6 | $0.048 | 8/22 | 8.42s |
| #126 | Gemini 3.1 Flash Lite minimal | 3 | 6.1 | $0.047 | 10/22 | 1.86s | |
| #176 | GLM 5 Turbo none | Z.ai | 2 | 5.1 | $0.047 | 6/21 | 2.82s |
| #128 | Gemini 3.1 Flash Lite none | 1 | 6.1 | $0.046 | 9/22 | 1.75s | |
| #153 | Mimo V2 PRO none | Xiaomi | 2 | 5.6 | $0.045 | 7/21 | 2.27s |
| #119 | MiMo-V2-Flash medium | Xiaomi | 1 | 6.3 | $0.043 | 12/21 | 20.1s |
| #156 | DeepSeek V4 Flash none | DeepSeek | 1 | 5.6 | $0.042 | 5/22 | 36.8s |
| #164 | KAT-Coder-Air V2.5 low | Kwaipilot | 2 | 5.4 | $0.041 | 7/22 | 10.1s |
| #186 | GPT-5.4 Nano none | OpenAI | 2 | 4.8 | $0.041 | 4/22 | 2.57s |
| #49 | DeepSeek V4 Flash high | DeepSeek | 2 | 7.7 | $0.041 | 13/22 | 49.7s |
| #210 | Qwen3.5-9B medium | Qwen | 1 | 3.8 | $0.036 | 3/22 | 82.2s |
| #192 | Laguna M.1 medium | Poolside | 1 | 4.7 | $0.033 | 9/19 | 14.7s |
| #193 | Qwen3 Coder Next medium | Qwen | 3 | 4.7 | $0.032 | 4/22 | 9.61s |