Instructions following Model Ranking

See which AI models perform best on Instructions following, which ones stay reliable, and where the biggest gaps appear. Sort by: Total Cost ↓.

Models Shown

Average Instructions following Score

0.8

Best Model

Claude Sonnet 5.5 0.9

Failure Reasons

With failure reason Wrong answer118 With failure reason Did not follow instructions82 With failure reason API error25 With failure reason Extra formatting8 With failure reason Timed out5 With failure reason No answer2

461/461

Rank	Model	Company	Instructions following Score	Score	Total Cost	Tests Correct	Response Time (avg)
#387	Nemotron 3.5 Lightning low	NVIDIA	0.8	0.5	$0.175	1/2	5.46s
Total Tests 2 Wrong Tests 1 Total Cost $0.175 Response Time (avg) 5.46s
#312	GPT-5.4 Mini none	OpenAI	0.6	0.6	$0.170	1/2	728ms
Total Tests 2 Wrong Tests 1 Total Cost $0.170 Response Time (avg) 728ms
#348	Nemotron 3.5 Lightning high	NVIDIA	0.8	0.5	$0.168	1/2	4.98s
Total Tests 2 Wrong Tests 1 Total Cost $0.168 Response Time (avg) 4.98s
#372	Hy4 preview none	Tencent	0.7	0.5	$0.165	1/2	47.9s
Total Tests 2 Wrong Tests 1 Total Cost $0.165 Response Time (avg) 47.9s
#302	Nemotron 3 Ultra none	NVIDIA	1.0	0.6	$0.164	2/2	1.46s
Total Tests 2 Wrong Tests 0 Total Cost $0.164 Response Time (avg) 1.46s
#311	GPT-5 Nano medium	OpenAI	1.0	0.6	$0.159	2/2	15.6s
Total Tests 2 Wrong Tests 0 Total Cost $0.159 Response Time (avg) 15.6s
#195	Gemini 3.1 Flash Lite Preview medium	Google	1.0	0.7	$0.158	2/2	1.91s
Total Tests 2 Wrong Tests 0 Total Cost $0.158 Response Time (avg) 1.91s
#293	GLM 5.3 FlashX low	Z.ai	1.0	0.6	$0.157	2/2	1.34s
Total Tests 2 Wrong Tests 0 Total Cost $0.157 Response Time (avg) 1.34s
#248	Gemini 3 Flash Preview none	Google	0.6	0.6	$0.155	1/2	1.58s
Total Tests 2 Wrong Tests 1 Total Cost $0.155 Response Time (avg) 1.58s
#204	Gemini 3.1 Flash Lite medium	Google	1.0	0.7	$0.153	2/2	2.59s
Total Tests 2 Wrong Tests 0 Total Cost $0.153 Response Time (avg) 2.59s
#272	DeepSeek V3.2 medium	DeepSeek	1.0	0.6	$0.152	2/2	35.8s
Total Tests 2 Wrong Tests 0 Total Cost $0.152 Response Time (avg) 35.8s
#274	DeepSeek V4 Flash 0423 none	DeepSeek	0.7	0.6	$0.149	1/2	17.5s
Total Tests 2 Wrong Tests 1 Total Cost $0.149 Response Time (avg) 17.5s
#115	DeepSeek V4 Pro 0423 high	DeepSeek	0.8	0.8	$0.149	1/2	8.73s
Total Tests 2 Wrong Tests 1 Total Cost $0.149 Response Time (avg) 8.73s
#262	Qwen3.6 27B none	Qwen	0.6	0.6	$0.148	1/2	1.92s
Total Tests 2 Wrong Tests 1 Total Cost $0.148 Response Time (avg) 1.92s
#367	Laguna S 2.1 high	Poolside	0.5	0.5	$0.145	0/2	453ms
Total Tests 2 Wrong Tests 2 Total Cost $0.145 Response Time (avg) 453ms

Instructions following Ranking

Filter models

Top Models by Instructions following Score

Instructions following Score vs Total Cost

Top Models by Response Time (avg)