Coding Model Ranking

AI BENCHY Category

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

Models Shown

Average Coding Score

6.1

Best Model

Failure Reasons

With failure reason Wrong answer120 With failure reason No answer18 With failure reason Did not follow instructions16 With failure reason Timed out12 With failure reason Extra formatting7 With failure reason API error6

Rank	Model	Company	Coding Score	Score	Tests Correct	Response Time (avg)
#117	Grok 4.20 Beta none	X AI	5.5	5.5	0/1	1.14s
#85	Gemini 3.1 Flash Lite none	Google	6.8	6.6	1/2	1.13s
#147	GPT-5.4 Nano none	OpenAI	5.4	4.5	0/2	1.09s
#49	Gemini 3.1 Flash Lite Preview none	Google	6.8	7.5	1/2	1.06s
#135	Mistral Small 4 none	Mistral	4.0	5.0	0/2	1.03s
#136	GPT-5.4 Mini none	OpenAI	6.8	4.9	1/2	1.01s
#97	Qwen3.5-Flash none	Qwen	6.8	5.9	1/2	993ms
#78	Gemini 3.1 Flash Lite minimal	Google	6.8	6.7	1/2	951ms
#143	Mercury 2 none	Inception	3.5	4.6	0/2	831ms
#92	Gemini 2.5 Flash none	Google	6.8	6.2	1/2	810ms
#153	Granite 4.1 8B none	IBM Granite	5.2	4.1	0/2	706ms
#16	Qwen3.6 Plus Preview medium	Qwen	0.0	8.2	0/0	0ms
#19	Gemini 3 PRO Preview medium	Google	3.0	8.1	0/2	0ms
#32	Step 3.5 Flash none	Stepfun	3.0	7.8	0/1	0ms
#75	Hunter Alpha medium	OpenRouter	3.0	6.7	0/1	0ms

←

1 9 10 11

→

Coding Ranking