AI BENCHY
AD
Track all your projects in one dashboard. Get 📊stats, 🔥heatmaps and 👀recordings in one self-hosted dashboard.
uxwizz.com

AI BENCHY Category Failures

Anti-AI Tricks: Did not follow instructions

Anti-AI Tricks
Did not follow instructions

See which AI models are most likely to hit Did not follow instructions on Anti-AI Tricks, so you can spot weak points faster. Sort by: Tests Correct ↓.

Models Shown

14

Total Failures

30

Most Affected Model

Step 3.7 Flash 1
Rank Model Company Did not follow instructions Count Category Score Tests Correct Response Time (avg)
#137 Elephant Alpha none Openrouter 1 6.6 2/4 963ms
#138 Ling-2.6-flash none Inclusionai 1 6.8 2/4 11.8s
#67 MiniMax M3 medium Minimax 1 5.5 1/4 14.9s
#156 Hy3 preview none Tencent 2 4.8 1/4 11.1s
#158 GLM 4.7 Flash medium Z.ai 1 4.7 1/4 15.0s
#163 Granite 4.1 8B none IBM Granite 1 4.9 1/4 844ms
#101 Mimo V2 Omni none Xiaomi 1 3.6 0/4 1.63s
#113 DeepSeek V4 Pro none DeepSeek 1 3.5 0/4 14.0s
#121 Owl Alpha none Openrouter 1 3.4 0/4 2.78s
#123 MiMo-V2.5-Pro none Xiaomi 1 3.3 0/4 2.67s
#140 Qwen3 Coder Next none Qwen 1 3.6 0/4 3.31s
#146 Laguna Xs.2 none Poolside 1 3.0 0/4 534ms
#150 Qwen3 Coder Next medium Qwen 1 3.5 0/4 8.64s
#157 Grok 4.1 Fast none X AI 1 3.2 0/4 1.07s

Top Models by Did not follow instructions Count

Did not follow instructions Count vs Score

Top Models by Response Time (avg)

Top Models by Estimated Wasted Cost