Combined Model Ranking

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

Models Shown

Average Combined Score

5.6

Best Model

Failure Reasons

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

216/216

Rank	Model	Company	Combined Score	Score	Total Cost	Tests Correct	Response Time (avg)
#143	North Mini Code medium	Cohere	2.9	5.9	$0.000	0/2	554.9s
Total Tests 2 Wrong Tests 2 Total Cost $0.000 Response Time (avg) 554.9s
#105	Qwen3.6 27B medium	Qwen	6.7	6.5	$0.779	0/2	584.1s
Total Tests 2 Wrong Tests 2 Total Cost $0.779 Response Time (avg) 584.1s
#62	Qwen3.5-27B medium	Qwen	7.3	7.4	$1.627	1/2	595.2s
Total Tests 2 Wrong Tests 1 Total Cost $1.627 Response Time (avg) 595.2s
#200	GLM 4.7 Flash medium	Z.ai	2.9	4.3	$0.166	0/2	802.8s
Total Tests 2 Wrong Tests 2 Total Cost $0.166 Response Time (avg) 802.8s
#136	Step 3.5 Flash medium	Stepfun	6.5	6.0	$0.108	1/2	813.7s
Total Tests 2 Wrong Tests 1 Total Cost $0.108 Response Time (avg) 813.7s
#94	Qwen3.6 35B A3B medium	Qwen	3.0	6.7	$0.746	0/2	817.6s
Total Tests 2 Wrong Tests 2 Total Cost $0.746 Response Time (avg) 817.6s

Combined Ranking