Did not follow instructions Failure Ranking

AI BENCHY Failures

See which AI models run into Did not follow instructions most often, so you can spot reliability risks before choosing one. Sort by: Response Time (avg) ↑.

Models Shown

Total Failures

180

Most Affected Model

Mercury 2 1

Categories

In category Puzzle Solving75 In category General Intelligence58 In category Instructions following22 In category Anti-AI Tricks17 In category Tool Calling6 In category Coding2

Rank	Model	Company	Did not follow instructions Count	Score	Tests Correct	Response Time (avg)
#32	Qwen3.5-Flash medium	Qwen	1	7.8	11/18	66.7s
#11	Gemini 3.1 Flash Lite Preview high	Google	1	8.4	12/16	68.8s
#39	Seed-2.0-Mini medium	Bytedance Seed	1	7.5	11/18	69.7s
#46	Kimi K2.5 medium	Moonshot AI	2	7.0	9/18	72.4s
#97	Qwen3.5-9B medium	Qwen	2	4.4	3/18	73.6s

←

1 2 3 4 5 6

Did not follow instructions Failures

Top Models by Did not follow instructions Count

Did not follow instructions Count vs Score

Top Models by Response Time (avg)