Combined Model Ranking

See which AI models perform best on Combined, which ones stay reliable, and where the biggest gaps appear. Sort by: Total Cost ↑.

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)
#169	Gemini 3.1 Flash Lite Preview high	Google	5.0	5.3	$2.310	1/1	280.5s
Total Tests 1 Wrong Tests 0 Total Cost $2.310 Response Time (avg) 280.5s
#47	Claude Opus 4.6 medium	Anthropic	10.0	7.7	$3.059	2/2	114.1s
Total Tests 2 Wrong Tests 0 Total Cost $3.059 Response Time (avg) 114.1s
#37	Kimi K3 max	Moonshot AI	6.5	8.0	$3.112	1/2	223.0s
Total Tests 2 Wrong Tests 1 Total Cost $3.112 Response Time (avg) 223.0s
#20	Claude Fable 5 medium	Anthropic	6.5	8.6	$3.478	1/2	27.5s
Total Tests 2 Wrong Tests 1 Total Cost $3.478 Response Time (avg) 27.5s
#13	GPT-5.5 medium	OpenAI	10.0	9.0	$4.137	2/2	33.5s
Total Tests 2 Wrong Tests 0 Total Cost $4.137 Response Time (avg) 33.5s
#187	Grok 4.20 Multi Agent Beta medium	X AI	1.5	4.8	$5.599	0/1	0ms
Total Tests 1 Wrong Tests 1 Total Cost $5.599 Response Time (avg) 0ms

Combined Ranking