Best AI Models for Coding

Q: Which AI model is best for coding?

The top-ranked coding model changes frequently. Check the live leaderboard above for the current leader based on HumanEval+, SWE-bench Verified, and other coding benchmarks.

Q: What benchmarks measure coding ability?

Key coding benchmarks include HumanEval+ (function completion), SWE-bench Verified (real GitHub issues), Aider Polyglot (multi-language editing), and Terminal-bench (CLI tasks).

Q: Are open-source models competitive for coding?

Yes. Several open-source models score competitively on coding benchmarks, especially models from DeepSeek and Meta.

AI models ranked by coding benchmarks. Compare HumanEval+, SWE-bench Verified, Aider Polyglot, and more across all providers.

Models

Providers

Open Source

$0.97

Median $/1M in

Top 3

Claude Mythos Preview

🇺🇸 Anthropic

1.0MN/A/M in

81.814 benchmarks

Full Rankings

#	Model	Avg	aider poly	GSO-Bench	swe bench	swe bench	terminal b	$/1M in	Context
1	GPT-5.5🇺🇸 OpenAI	85.0	-	-	-	-	82.7	$5.00	400K
2	GPT-5 Chat🇺🇸 OpenAI	81.9	88.0	-	-	-	-	$1.25	128K
3	Claude Mythos Preview🇺🇸 Anthropic	81.8	-	-	93.9	-	82.0	N/A	1.0M
4	Gemini 2.5 Pro Preview 05-06🇺🇸 Google DeepMind	76.9	76.9	-	-	-	-	$1.25	1.0M
5	o4 Mini High🇺🇸 OpenAI	72.0	72.0	-	-	-	-	$1.10	200K
6	Grok 3 Beta🇺🇸 xAI	69.5	53.3	-	-	-	-	$3.00	131K
7	gpt-oss-120b (free)🇺🇸 OpenAIOpen	68.7	41.8	-	-	-	-	Free	131K
8	Grok 4.20🇺🇸 xAI	66.1	-	-	-	-	57.3	$1.25	2.0M
9	Grok 3 Mini Beta🇺🇸 xAI	64.8	49.3	-	-	-	-	$0.30	131K
10	Gemini 3.5 Flash🇺🇸 Google DeepMind	62.6	-	-	79.3	-	-	$1.50	1.0M
11	o3 Pro🇺🇸 OpenAI	61.2	84.9	-	-	-	-	$20.00	200K
12	Gemini 3 Pro🇺🇸 Google DeepMind	60.4	-	18.6	72.9	-	69.4	N/A	0K
13	o3 Mini High🇺🇸 OpenAI	60.4	60.4	-	-	-	-	$1.10	200K
14	Qwen3.6 Plus🇨🇳 Alibaba QwenOpen	59.9	-	-	57.9	-	-	$0.33	1.0M
15	GLM 5.1🇨🇳 z-aiOpen	59.8	-	-	74.2	-	-	$0.97	203K
16	DeepSeek V3🇨🇳 DeepSeekOpen	59.0	48.4	-	-	-	-	$0.20	131K
17	GPT-5.4🇺🇸 OpenAI	58.9	-	31.4	76.9	-	81.8	$2.50	1.1M
18	Claude Opus 4.7🇺🇸 Anthropic	58.2	-	44.1	83.5	-	90.2	$5.00	1.0M
19	Qwen3 32B🇨🇳 Alibaba QwenOpen	58.2	40.0	-	-	-	-	$0.08	131K
20	R1 0528🇨🇳 DeepSeekOpen	57.9	71.4	-	-	-	-	$0.50	164K
21	Claude Opus 4.6🇺🇸 Anthropic	57.1	-	41.2	78.7	-	79.8	$5.00	1.0M
22	Gemini 3.1 Pro Preview🇺🇸 Google DeepMind	56.6	-	22.6	75.6	-	80.2	$2.00	1.0M
23	Qwen3 235B A22B🇨🇳 Alibaba QwenOpen	56.4	59.6	-	-	-	-	$0.46	131K
24	Kimi K2 0711🇨🇳 moonshotaiOpen	56.2	59.1	4.9	-	-	27.8	$0.57	131K
25	GLM 5🇨🇳 z-aiOpen	56.1	-	-	72.1	-	52.4	$0.60	203K
26	DeepSeek V3 0324🇨🇳 DeepSeekOpen	55.1	55.1	-	-	-	-	$0.24	164K
27	MiniMax M2.5🇨🇳 minimaxOpen	55.1	-	-	-	-	42.7	$0.15	205K
28	Grok 4🇺🇸 xAI	54.8	79.6	-	-	-	27.2	N/A	0K
29	GPT-5 Mini🇺🇸 OpenAI	54.3	-	-	64.7	59.8	34.8	$0.25	400K
30	o3🇺🇸 OpenAI	53.9	81.3	8.8	62.3	58.4	-	$2.00	200K
31	Gemini 2.0 Pro🇺🇸 Google DeepMind	53.7	35.6	-	-	-	-	N/A	0K
32	GPT-5.2🇺🇸 OpenAI	53.6	-	27.4	73.8	71.8	64.9	$1.75	400K
33	Grok-3 mini🇺🇸 xAI	53.3	49.3	-	-	-	-	N/A	0K
34	Kimi K2 Thinking🇨🇳 moonshotaiOpen	53.3	-	-	-	63.4	35.7	$0.60	262K
35	DeepSeek V3.2 Exp🇨🇳 DeepSeekOpen	53.2	74.2	-	-	-	-	$0.27	164K
36	GPT-5🇺🇸 OpenAI	53.2	88.0	6.9	73.5	65.0	49.6	$1.25	400K
37	Qwen2.5 Coder 32B Instruct🇨🇳 Alibaba QwenOpen	53.1	16.4	-	-	-	-	$0.66	128K
38	Gemini 2.5 Pro🇺🇸 Google DeepMind	52.7	83.1	3.9	57.6	-	32.6	$1.25	1.0M
39	GPT-5.3-Codex🇺🇸 OpenAI	52.4	-	-	74.8	-	78.4	$1.75	400K
40	Kimi K2.5🇨🇳 moonshotaiOpen	52.0	-	-	73.8	-	43.2	$0.38	262K
41	Kimi K2.6🇨🇳 moonshotaiOpen	51.7	-	-	76.7	-	-	$0.66	262K
42	DeepSeek V3.2🇨🇳 DeepSeekOpen	51.2	74.2	-	-	-	39.6	$0.21	131K
43	Qwen3.6 Max Preview🇨🇳 Alibaba QwenOpen	51.2	-	-	76.7	-	-	$1.04	262K
44	Gemini 2.5 Pro Preview 06-05🇺🇸 Google DeepMind	50.9	83.1	-	-	-	-	$1.25	1.0M
45	o4 Mini🇺🇸 OpenAI	50.9	72.0	3.6	-	45.0	-	$1.10	200K
46	GLM 4.6🇨🇳 z-aiOpen	50.8	-	-	-	-	24.5	$0.43	203K
47	GLM 4.7🇨🇳 z-aiOpen	50.5	-	-	-	-	33.4	$0.40	203K
48	Grok 4 Fast🇺🇸 xAI	50.4	-	-	-	-	-	N/A	0K
49	GPT-5.1🇺🇸 OpenAI	49.6	-	13.7	68.0	66.0	47.6	$1.25	400K
50	o1🇺🇸 OpenAI	49.6	61.7	-	-	-	-	$15.00	200K
51	Qwen3 235B A22B Instruct 2507🇨🇳 Alibaba QwenOpen	48.5	59.6	-	-	-	-	$0.09	262K
52	Claude Sonnet 4.6🇺🇸 Anthropic	48.0	-	-	75.2	-	53.4	$3.00	1.0M
53	Gemini 2.0 Flash🇺🇸 Google DeepMind	48.0	38.2	-	-	-	-	$0.10	1.0M
54	Gemini 2.0 Flash (Dec 2024)🇺🇸 Google DeepMind	48.0	38.2	-	-	-	-	N/A	0K
55	Claude 3.7 Sonnet🇺🇸 Anthropic	47.7	64.9	3.8	61.0	52.8	-	N/A	0K
56	Gemini 3 Flash Preview🇺🇸 Google DeepMind	47.7	-	9.8	75.4	-	64.3	$0.50	1.0M
57	gpt-oss-120b🇺🇸 OpenAIOpen	46.9	41.8	-	-	26.0	18.7	$0.04	131K
58	Grok 3🇺🇸 xAI	45.5	53.3	-	-	-	-	N/A	0K
59	R1🇨🇳 DeepSeekOpen	45.1	56.9	-	-	-	-	$0.70	164K
60	GPT-4.1 Mini🇺🇸 OpenAI	44.5	32.4	-	-	23.9	-	$0.40	1.0M
61	Claude Opus 4.5🇺🇸 Anthropic	43.6	-	26.5	76.7	74.4	63.1	$5.00	200K
62	Claude Sonnet 4🇺🇸 Anthropic	43.3	61.3	4.9	-	64.9	-	$3.00	1.0M
63	GPT-4.1🇺🇸 OpenAI	43.3	52.4	-	48.5	39.6	-	$2.00	1.0M
64	GPT-5 Nano🇺🇸 OpenAI	43.3	-	-	-	34.8	21.8	$0.05	400K
65	GPT-4o-mini (2024-07-18)🇺🇸 OpenAI	43.2	3.6	-	-	-	-	$0.15	128K
66	Claude 3.5 Sonnet🇺🇸 Anthropic	42.3	51.6	4.6	-	-	-	N/A	0K
67	Gemma 3 27B🇺🇸 Google DeepMindOpen	42.2	4.9	-	-	-	-	$0.08	131K
68	Gemma 3 27B (free)🇺🇸 Google DeepMindOpen	42.2	4.9	-	-	-	-	Free	131K
69	Claude Opus 4🇺🇸 Anthropic	41.7	72.0	6.9	70.7	67.6	-	$15.00	200K
70	Gemini 1.5 Pro (May 2024)🇺🇸 Google DeepMind	41.5	-	-	-	-	-	N/A	0K
71	o1-preview🇺🇸 OpenAI	41.5	-	-	-	-	-	N/A	0K
72	Gemini 1.5 Pro (Feb 2024)🇺🇸 Google DeepMind	41.3	-	-	-	-	-	N/A	0K
73	Qwen2.5-Max🇨🇳 Alibaba QwenOpen	41.0	21.8	-	-	-	-	N/A	0K
74	Claude Sonnet 4.5🇺🇸 Anthropic	39.6	-	14.7	71.3	70.6	46.5	$3.00	1.0M
75	GPT-4o-mini🇺🇸 OpenAI	39.6	3.6	-	-	-	-	$0.15	128K
76	Gemini 2.5 Flash🇺🇸 Google DeepMind	38.5	47.1	-	-	-	17.1	$0.30	1.0M
77	Llama 3.1 405B🇺🇸 MetaOpen	38.0	-	-	-	-	-	N/A	0K
78	Gemini 2.0 Flash Thinking (Jan 2025)🇺🇸 Google DeepMind	37.7	18.2	-	-	-	-	N/A	0K
79	Claude 3.5 Haiku🇺🇸 Anthropic	37.2	28.0	-	-	-	-	N/A	0K
80	Claude Opus 4.1🇺🇸 Anthropic	37.0	-	-	73.3	-	38.0	$15.00	200K
81	GPT-4o (2024-11-20)🇺🇸 OpenAI	36.4	23.1	0.0	31.0	21.6	-	$2.50	128K
82	GPT-4.5🇺🇸 OpenAI	35.9	44.9	-	-	-	-	N/A	0K
83	GPT-4o (2024-08-06)🇺🇸 OpenAI	35.6	23.1	-	-	-	-	$2.50	128K
84	GPT-4.1 Nano🇺🇸 OpenAI	35.2	8.9	-	-	-	-	$0.10	1.0M
85	o3 Mini🇺🇸 OpenAI	35.2	60.4	1.3	-	-	-	$1.10	200K
86	o1-mini🇺🇸 OpenAI	34.9	32.9	-	-	-	-	N/A	0K
87	Claude Haiku 4.5🇺🇸 Anthropic	34.3	-	-	-	-	35.5	$1.00	200K
88	Claude 3 Opus🇺🇸 Anthropic	33.7	-	-	-	-	-	N/A	0K
89	Llama 3 70B Instruct🇺🇸 MetaOpen	32.4	-	-	-	-	-	$0.51	8K
90	Claude 3 Haiku🇺🇸 Anthropic	28.7	-	-	-	-	-	$0.25	200K
91	Llama 4 Maverick🇺🇸 MetaOpen	28.0	15.6	-	-	21.0	-	$0.15	1.0M
92	Llama 3-70B🇺🇸 MetaOpen	24.7	-	-	-	-	-	N/A	0K
93	Mixtral 8x22B Instruct🇫🇷 Mistral AIOpen	23.5	-	-	-	-	-	$2.00	66K
94	Llama 4 Scout🇺🇸 MetaOpen	18.9	-	-	-	9.1	-	$0.10	10.0M
95	QwQ 32B🇨🇳 Alibaba QwenOpen	13.5	20.9	-	-	-	-	$0.15	131K

90+ Gold 80-89 70-79 60-69 <60Scores in % unless noted. Avg = unweighted mean across tested benchmarks.

About this category

Models ranked by coding performance across HumanEval+, SWE-bench Verified, Aider Polyglot, and other code generation benchmarks. Scores reflect real-world software engineering tasks including bug fixing, code completion, and multi-file editing.

Related categories

Best AI Models for Reasoning

AI models ranked by reasoning benchmarks. Compare GPQA Diamond, ARC-AGI, BBH, and other reasoning tests across all providers.

Best AI Models for Math

AI models ranked by math benchmarks. Compare MATH-500, GSM8K, and competition-level math scores across all providers.

Flagship AI Models

The best AI model from each provider, ranked by benchmark score. Compare the flagships from OpenAI, Anthropic, Google, Meta, and more.

Frequently asked questions

Which AI model is best for coding?

The top-ranked coding model changes frequently. Check the live leaderboard above for the current leader based on HumanEval+, SWE-bench Verified, and other coding benchmarks.

What benchmarks measure coding ability?

Key coding benchmarks include HumanEval+ (function completion), SWE-bench Verified (real GitHub issues), Aider Polyglot (multi-language editing), and Terminal-bench (CLI tasks).

Are open-source models competitive for coding?

Yes. Several open-source models score competitively on coding benchmarks, especially models from DeepSeek and Meta.

Back to all models