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: Failure Count ↑.

Models Shown

Total Failures

215

Most Affected Model

Categories

In category Puzzle Solving78 In category General Intelligence74 In category Anti-AI Tricks30 In category Coding14 In category Instructions following11 In category Tool Calling6 In category Combined1 In category Domain specific1

Rank	Model	Company	Did not follow instructions Count	Score	Tests Correct	Response Time (avg)
#65	Grok 4.20 medium	X AI	2	7.1	12/21	27.7s
#67	MiniMax M3 medium	Minimax	2	7.1	11/21	68.2s
#70	GPT-5.4 Nano medium	OpenAI	2	7.0	11/21	12.0s
#75	Ring-2.6-1T medium	Inclusionai	2	6.9	11/21	61.3s
#76	Kimi K2.5 medium	Moonshot AI	2	6.8	10/21	98.4s
#79	Hunter Alpha medium	OpenRouter	2	6.7	8/18	10.3s
#80	Mimo V2 Omni medium	Xiaomi	2	6.7	10/21	41.2s
#84	Grok 4.20 Multi Agent Beta medium	X AI	2	6.6	8/18	9.69s
#94	GPT-5 Nano medium	OpenAI	2	6.3	9/21	42.5s
#96	Ring-2.6-1T none	Inclusionai	2	6.2	9/21	55.1s
#100	Grok Build 0.1 none	X AI	2	6.0	7/19	28.7s
#102	Gemma 4 26B A4B none	Google	2	6.0	8/21	5.91s
#109	GLM 5V Turbo none	Z.ai	2	5.8	8/21	2.99s
#111	Owl Alpha medium	Openrouter	2	5.7	8/21	11.9s
#113	DeepSeek V4 Pro none	DeepSeek	2	5.7	7/21	12.4s

←

1 4 5 6 8

→

Did not follow instructions Failures