Anti-AI Tricks Model Ranking

See which AI models perform best on Anti-AI Tricks, which ones stay reliable, and where the biggest gaps appear. Sort by: Response Time (avg) ↑.

Models Shown

Average Anti-AI Tricks Score

7.2

Best Model

Failure Reasons

With failure reason Wrong answer293 With failure reason Did not follow instructions33 With failure reason Extra formatting20 With failure reason API error14 With failure reason No answer4 With failure reason Timed out4

216/216

Rank	Model	Company	Anti-AI Tricks Score	Score	Total Cost	Tests Correct	Response Time (avg)
#169	Gemini 3.1 Flash Lite Preview high	Google	7.5	5.3	$2.310	3/3	43.9s
Total Tests 3 Wrong Tests 0 Total Cost $2.310 Response Time (avg) 43.9s
#61	Qwen3.5 Plus 2026-02-15 medium	Qwen	8.2	7.5	$0.437	3/4	45.8s
Total Tests 4 Wrong Tests 1 Total Cost $0.437 Response Time (avg) 45.8s
#81	Kimi K2.5 medium	Moonshot AI	7.3	7.0	$0.600	2/4	51.4s
Total Tests 4 Wrong Tests 2 Total Cost $0.600 Response Time (avg) 51.4s
#120	Qwen3.5-Flash medium	Qwen	10.0	6.2	$0.139	4/4	59.1s
Total Tests 4 Wrong Tests 0 Total Cost $0.139 Response Time (avg) 59.1s
#143	North Mini Code medium	Cohere	8.4	5.9	$0.000	3/4	64.8s
Total Tests 4 Wrong Tests 1 Total Cost $0.000 Response Time (avg) 64.8s
#84	Seed-2.0-Mini medium	Bytedance Seed	6.6	7.0	$0.101	2/4	74.7s
Total Tests 4 Wrong Tests 2 Total Cost $0.101 Response Time (avg) 74.7s

Anti-AI Tricks Ranking