Combined x Wrong answer Ranking

See which AI models are most likely to hit Wrong answer on Combined, so you can spot weak points faster. Sort by: Tests Correct ↑.

Models Shown

Total Failures

Most Affected Model

Failure Reasons

Invalid tool call91 Wrong answer69 No answer32 API error26 Timed out5 Did not follow instructions1 Extra formatting1

Categories

Domain specific421 Anti-AI Tricks293 Coding259 Puzzle Solving204 Trivia172 Combined69 General Intelligence62 Instructions following61 Data parsing and extraction41 Tool Calling3

64/64

Rank	Model	Company	Wrong answer Count	Category Score	Total Cost	Tests Correct	Response Time (avg)
#117	LongCat 2.0 none	Meituan	1	6.5	$0.044	1/2	28.4s
Total Tests 2 Wrong Tests 1 Total Cost $0.044 Response Time (avg) 28.4s
#132	Qwen3.5 Plus 2026-04-20 none	Qwen	1	6.4	$0.122	1/2	109.7s
Total Tests 2 Wrong Tests 1 Total Cost $0.122 Response Time (avg) 109.7s
#142	GPT-5.4 Mini none	OpenAI	1	6.5	$0.095	1/2	6.22s
Total Tests 2 Wrong Tests 1 Total Cost $0.095 Response Time (avg) 6.22s
#168	Ling-2.6-1T none	Inclusionai	1	6.5	$0.016	1/2	23.8s
Total Tests 2 Wrong Tests 1 Total Cost $0.016 Response Time (avg) 23.8s

←

1 2 3 4 5

Filter models