Anti-AI Tricks x Wrong answer Ranking

See which AI models are most likely to hit Wrong answer on Anti-AI Tricks, so you can spot weak points faster. Sort by: Total Cost ↑.

Models Shown

Total Failures

293

Most Affected Model

Failure Reasons

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

Categories

Domain specific412 Anti-AI Tricks293 Coding252 Puzzle Solving201 Trivia168 Combined68 Instructions following61 General Intelligence59 Data parsing and extraction41 Tool Calling3

140/140

Rank	Model	Company	Wrong answer Count	Category Score	Total Cost	Tests Correct	Response Time (avg)
#18	GPT-5.4 medium	OpenAI	1	8.3	$1.533	3/4	4.11s
Total Tests 4 Wrong Tests 1 Total Cost $1.533 Response Time (avg) 4.11s
#27	Muse Spark 1.1 high	Meta	1	7.5	$1.694	2/4	8.60s
Total Tests 4 Wrong Tests 2 Total Cost $1.694 Response Time (avg) 8.60s
#143	Gemini 3.1 Flash Lite high	Google	1	8.7	$2.044	3/4	37.2s
Total Tests 4 Wrong Tests 1 Total Cost $2.044 Response Time (avg) 37.2s
#40	Claude Sonnet 4.6 medium	Anthropic	1	6.5	$2.057	2/4	2.98s
Total Tests 4 Wrong Tests 2 Total Cost $2.057 Response Time (avg) 2.98s
#181	Grok 4.20 Multi Agent Beta medium	X AI	1	6.9	$5.599	2/4	3.46s
Total Tests 4 Wrong Tests 2 Total Cost $5.599 Response Time (avg) 3.46s

Filter models