Combined Model Ranking

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

Models Shown

Average Combined Score

0.6

Best Model

Muse Glimmer 30B 0.6

Failure Reasons

With failure reason Invalid tool call180 With failure reason Wrong answer128 With failure reason API error62 With failure reason No answer56 With failure reason Timed out11 With failure reason Did not follow instructions1 With failure reason Extra formatting1

451/451

Rank	Model	Company	Combined Score	Score	Total Cost	Tests Correct	Response Time (avg)
#122	Qwen3.8 2.4T A95B high	Qwen	0.7	0.8	$2.949	1/2	266.1s
Total Tests 2 Wrong Tests 1 Total Cost $2.949 Response Time (avg) 266.1s
#123	MiMo-V2.6-Flash medium	Xiaomi	0.7	0.8	$0.218	1/2	20.7s
Total Tests 2 Wrong Tests 1 Total Cost $0.218 Response Time (avg) 20.7s
#125	Ling 3.1 Flash high	Inclusionai	0.6	0.8	$0.000	1/2	351.2s
Total Tests 2 Wrong Tests 1 Total Cost $0.000 Response Time (avg) 351.2s
#128	GLM 5.3 high	Z.ai	0.6	0.8	$0.495	1/2	64.0s
Total Tests 2 Wrong Tests 1 Total Cost $0.495 Response Time (avg) 64.0s
#129	Step 3.7 Flash medium	Stepfun	0.7	0.8	$0.571	1/2	80.9s
Total Tests 2 Wrong Tests 1 Total Cost $0.571 Response Time (avg) 80.9s
#132	Grok 4.5 low	X AI	0.7	0.8	$1.345	1/2	12.8s
Total Tests 2 Wrong Tests 1 Total Cost $1.345 Response Time (avg) 12.8s
#133	DeepSeek V4 Flash 0731 medium	DeepSeek	0.7	0.8	$0.460	1/2	150.6s
Total Tests 2 Wrong Tests 1 Total Cost $0.460 Response Time (avg) 150.6s
#136	Muse Glimmer 30B high	Meta	0.6	0.8	$0.506	1/2	115.4s
Total Tests 2 Wrong Tests 1 Total Cost $0.506 Response Time (avg) 115.4s
#137	Nemotron 3.5 Super VL medium	NVIDIA	0.6	0.8	$0.437	1/2	112.2s
Total Tests 2 Wrong Tests 1 Total Cost $0.437 Response Time (avg) 112.2s
#138	Claude Haiku 5.5 medium	Anthropic	0.7	0.8	$0.063	1/2	20.3s
Total Tests 2 Wrong Tests 1 Total Cost $0.063 Response Time (avg) 20.3s
#139	Qwen3.5 Plus 2026-04-20 medium	Qwen	0.7	0.8	$0.435	1/2	92.4s
Total Tests 2 Wrong Tests 1 Total Cost $0.435 Response Time (avg) 92.4s
#140	Qwen3.7 Flash low	Qwen	0.6	0.8	$0.071	1/2	217.8s
Total Tests 2 Wrong Tests 1 Total Cost $0.071 Response Time (avg) 217.8s
#141	Muse Glimmer 30B medium	Meta	0.7	0.8	$0.288	1/2	76.5s
Total Tests 2 Wrong Tests 1 Total Cost $0.288 Response Time (avg) 76.5s
#143	Qwen3.8 Flash Next medium	Qwen	0.6	0.8	~$0.047	1/2	76.7s
Total Tests 2 Wrong Tests 1 Total Cost ~$0.047 Response Time (avg) 76.7s
#144	GPT-6 Luna low	OpenAI	0.7	0.8	$0.033	1/2	49.8s
Total Tests 2 Wrong Tests 1 Total Cost $0.033 Response Time (avg) 49.8s

Combined Ranking

Filter models

Top Models by Combined Score

Combined Score vs Total Cost

Top Models by Response Time (avg)

Combined Ranking

Filter models

Top Models by Combined Score

Combined Score vs Total CostEstimated GPU electricity cost only. It excludes the rest of the computer and hardware ownership. NVIDIA GeForce RTX 3090: 0.4752 h × 0.300 kW × €0.2896/kWh × 1.138 USD/EUR = ~$0.046984.

Top Models by Response Time (avg)

Combined Score vs Total Cost