Puzzle Solving Model Ranking

See which AI models perform best on Puzzle Solving, which ones stay reliable, and where the biggest gaps appear. Sort by: Metric ↑.

Models Shown

Average Puzzle Solving Score

0.7

Best Model

Failure Reasons

With failure reason Wrong answer344 With failure reason Did not follow instructions160 With failure reason API error35 With failure reason Extra formatting16 With failure reason No answer14 With failure reason Timed out11

442/442

Rank	Model	Company	Puzzle Solving Score	Score	Total Cost	Tests Correct	Response Time (avg)
#296	Seed 2.1 Turbo none	Bytedance Seed	1.0	0.6	$0.182	3/3	2.31s
Total Tests 3 Wrong Tests 0 Total Cost $0.182 Response Time (avg) 2.31s
#314	Claude Opus 4.7 none	Anthropic	1.0	0.5	$0.505	3/3	2.46s
Total Tests 3 Wrong Tests 0 Total Cost $0.505 Response Time (avg) 2.46s
#320	Ember-1 none	Fireworks	1.0	0.5	$0.786	3/3	2.54s
Total Tests 3 Wrong Tests 0 Total Cost $0.786 Response Time (avg) 2.54s
#325	Qwen3.6 Max Preview none	Qwen	1.0	0.5	$0.228	3/3	2.65s
Total Tests 3 Wrong Tests 0 Total Cost $0.228 Response Time (avg) 2.65s
#346	Grok 4.20 Beta medium	X AI	1.0	0.5	$0.750	3/3	3.52s
Total Tests 3 Wrong Tests 0 Total Cost $0.750 Response Time (avg) 3.52s
#350	Gemini 3 PRO Preview medium	Google	1.0	0.5	$0.385	3/3	3.88s
Total Tests 3 Wrong Tests 0 Total Cost $0.385 Response Time (avg) 3.88s
#376	Mimo V2 Omni default	Xiaomi	1.0	0.4	$0.021	3/3	1.16s
Total Tests 3 Wrong Tests 0 Total Cost $0.021 Response Time (avg) 1.16s

Puzzle Solving Ranking