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: Tests Correct ↓.

Models Shown

Average Instructions following Score

8.6

Best Model

Gemini 3.6 Flash 9.9

Failure Reasons

With failure reason Wrong answer61 With failure reason Did not follow instructions19 With failure reason Extra formatting3 With failure reason No answer2 With failure reason API error1 With failure reason Timed out1

216/216

Rank	Model	Company	Instructions following Score	Score	Total Cost	Tests Correct	Response Time (avg)
#51	MiniMax M3 medium	Minimax	9.8	7.6	$0.286	2/2	6.14s
Total Tests 2 Wrong Tests 0 Total Cost $0.286 Response Time (avg) 6.14s
#52	Grok Build 0.1 medium	X AI	9.8	7.6	$1.097	2/2	12.4s
Total Tests 2 Wrong Tests 0 Total Cost $1.097 Response Time (avg) 12.4s
#53	GLM 5 Turbo medium	Z.ai	10.0	7.6	$0.323	2/2	5.38s
Total Tests 2 Wrong Tests 0 Total Cost $0.323 Response Time (avg) 5.38s
#54	GPT-5.6 Luna medium	OpenAI	9.9	7.6	$0.352	2/2	2.38s
Total Tests 2 Wrong Tests 0 Total Cost $0.352 Response Time (avg) 2.38s
#55	Nemotron 3 Ultra medium	NVIDIA	9.8	7.5	$0.774	2/2	6.35s
Total Tests 2 Wrong Tests 0 Total Cost $0.774 Response Time (avg) 6.35s
#56	Kimi K2.7 Code medium	Moonshot AI	9.9	7.5	$0.740	2/2	5.39s
Total Tests 2 Wrong Tests 0 Total Cost $0.740 Response Time (avg) 5.39s
#57	GPT-5.4 Nano medium	OpenAI	9.8	7.5	$0.138	2/2	1.88s
Total Tests 2 Wrong Tests 0 Total Cost $0.138 Response Time (avg) 1.88s
#58	GPT-5.3 Chat none	OpenAI	9.8	7.5	$0.571	2/2	3.51s
Total Tests 2 Wrong Tests 0 Total Cost $0.571 Response Time (avg) 3.51s
#59	GPT-5.6 Terra low	OpenAI	10.0	7.5	$0.519	2/2	1.48s
Total Tests 2 Wrong Tests 0 Total Cost $0.519 Response Time (avg) 1.48s
#60	GPT-5.4 Mini medium	OpenAI	9.8	7.5	$0.756	2/2	2.13s
Total Tests 2 Wrong Tests 0 Total Cost $0.756 Response Time (avg) 2.13s
#61	Qwen3.5 Plus 2026-02-15 medium	Qwen	10.0	7.5	$0.437	2/2	31.9s
Total Tests 2 Wrong Tests 0 Total Cost $0.437 Response Time (avg) 31.9s
#62	Qwen3.5-27B medium	Qwen	10.0	7.4	$1.627	2/2	19.7s
Total Tests 2 Wrong Tests 0 Total Cost $1.627 Response Time (avg) 19.7s
#63	Qwen3.7 Max none	Qwen	10.0	7.4	$0.197	2/2	943ms
Total Tests 2 Wrong Tests 0 Total Cost $0.197 Response Time (avg) 943ms
#65	Gemini 3 Flash Preview low	Google	9.9	7.4	$0.177	2/2	7.02s
Total Tests 2 Wrong Tests 0 Total Cost $0.177 Response Time (avg) 7.02s
#66	KAT-Coder-Pro V2.5 low	Kwaipilot	10.0	7.4	$0.387	2/2	2.53s
Total Tests 2 Wrong Tests 0 Total Cost $0.387 Response Time (avg) 2.53s

Instructions following Ranking

Filter models

Top Models by Instructions following Score

Instructions following Score vs Total Cost

Top Models by Response Time (avg)