AI BENCHY Failures
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.
131/131
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) |
|---|---|---|---|---|---|---|---|
| #161 | MiMo-V2.5 none | Xiaomi | 1 | 5.1 | $0.006 | 5/21 | 2.20s |
| #165 | Laguna M.1 medium | Poolside | 1 | 5.0 | $0.033 | 9/19 | 14.7s |
| #166 | GPT-4o-mini none | OpenAI | 1 | 5.0 | $0.006 | 5/21 | 1.77s |
| #169 | GLM 4.7 Flash none | Z.ai | 1 | 4.9 | $0.004 | 6/21 | 2.86s |
| #177 | Grok 4.20 Beta none | X AI | 1 | 4.7 | $0.087 | 6/18 | 1.19s |
| #179 | Mercury 2 none | Inception | 1 | 4.6 | $0.011 | 4/21 | 653ms |
| #190 | Laguna Xs.2 none | Poolside | 1 | 4.0 | $0.004 | 5/19 | 806ms |
| #193 | Qwen3.5-9B medium | Qwen | 1 | 3.8 | $0.036 | 3/21 | 82.2s |
| #194 | Nemotron 3 Nano Omni 30b A3b Reasoning medium | NVIDIA | 1 | 3.6 | $0.000 | 4/19 | 17.1s |
| #196 | Step 3.5 Flash none | Stepfun | 1 | 2.6 | $0.020 | 6/12 | 39.0s |
| #197 | LFM2-24B-A2B none | Liquid | 1 | 2.4 | $0.001 | 2/16 | 782ms |