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: Metric ↑.

Models Shown

Average Instructions following Score

0.8

Best Model

Gemini 3 Flash Preview 0.0

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)
#44	Claude Haiku 5.5 xhigh	Anthropic	1.0	0.9	$0.140	2/2	3.34s
Total Tests 2 Wrong Tests 0 Total Cost $0.140 Response Time (avg) 3.34s
#47	Pareto default	Unbiased	1.0	0.9	$1.520	2/2	4.95s
Total Tests 2 Wrong Tests 0 Total Cost $1.520 Response Time (avg) 4.95s
#50	Muse Spark Open xhigh	Meta	1.0	0.9	$2.081	2/2	23.5s
Total Tests 2 Wrong Tests 0 Total Cost $2.081 Response Time (avg) 23.5s
#52	Qwen3.8 Max (0902) medium	Qwen	1.0	0.9	$1.096	2/2	4.24s
Total Tests 2 Wrong Tests 0 Total Cost $1.096 Response Time (avg) 4.24s
#55	Gemini 4 Argon medium	Google	1.0	0.9	$0.674	2/2	2.71s
Total Tests 2 Wrong Tests 0 Total Cost $0.674 Response Time (avg) 2.71s
#56	Claude Opus 5 low	Anthropic	1.0	0.9	$2.350	2/2	4.13s
Total Tests 2 Wrong Tests 0 Total Cost $2.350 Response Time (avg) 4.13s
#63	Claude Fable 5 medium	Anthropic	1.0	0.9	$5.089	2/2	5.90s
Total Tests 2 Wrong Tests 0 Total Cost $5.089 Response Time (avg) 5.90s
#65	Kimi K3.1 Pro low	Moonshot AI	1.0	0.9	$1.836	2/2	5.84s
Total Tests 2 Wrong Tests 0 Total Cost $1.836 Response Time (avg) 5.84s
#66	Qwen3.8 MAX low	Qwen	1.0	0.9	$1.089	2/2	4.01s
Total Tests 2 Wrong Tests 0 Total Cost $1.089 Response Time (avg) 4.01s
#67	GPT-5.4 medium	OpenAI	1.0	0.9	$2.213	2/2	3.11s
Total Tests 2 Wrong Tests 0 Total Cost $2.213 Response Time (avg) 3.11s
#69	DeepSeek V4.1 Flash high	DeepSeek	1.0	0.9	$0.617	2/2	7.06s
Total Tests 2 Wrong Tests 0 Total Cost $0.617 Response Time (avg) 7.06s
#70	DeepSeek V4.1 Flash medium	DeepSeek	1.0	0.9	$0.591	2/2	9.69s
Total Tests 2 Wrong Tests 0 Total Cost $0.591 Response Time (avg) 9.69s
#71	Kimi K3.1 Pro max	Moonshot AI	1.0	0.9	$3.578	2/2	9.06s
Total Tests 2 Wrong Tests 0 Total Cost $3.578 Response Time (avg) 9.06s
#77	GPT-6 Sol low	OpenAI	1.0	0.9	$0.477	2/2	2.37s
Total Tests 2 Wrong Tests 0 Total Cost $0.477 Response Time (avg) 2.37s
#78	DeepSeek V4.1 Flash max	DeepSeek	1.0	0.9	$1.204	2/2	7.64s
Total Tests 2 Wrong Tests 0 Total Cost $1.204 Response Time (avg) 7.64s

Instructions following Ranking

Filter models

Top Models by Instructions following Score

Instructions following Score vs Total Cost

Top Models by Response Time (avg)