AA Coding Index leaderboard

A display-only Artificial Analysis coding index.

Models ranked by their published Artificial Analysis Coding Index value

Rank	Model	Provider	Aggregated model score
1	Claude Fable 5.1	Anthropic	81.6
2	Claude Opus 5	Anthropic	78.0
3	GPT-5.6 Sol	OpenAI	77.4
4	GPT-6 Astra	OpenAI	76.9
5	Grok 4.6	xAI	76.8
6	GPT-5.6 Terra	OpenAI	76.7
7	Claude Fable 5	Anthropic	76.5
8	Gemini 3.8 Flash	Google	76.3
9	Kimi K3	Moonshot AI	76.2
10	Gemini 3.7 Flash	Google	76.1
11	Muse Spark 1.3	Meta	75.8
12	GPT-5.5	OpenAI	74.9
13	GLM-5.3	Z.AI	74.8
14	Claude Opus 4.8	Anthropic	74.3
15	Claude Opus 4.7 (Adaptive)	Anthropic	73.6
16	Qwen3.8-Flash-Next	Alibaba	73.0
17	Grok 4.5	xAI	72.5
18	Muse Spark 1.2	Meta	72.2
19	Qwen3.8 Max Preview	Alibaba	71.8
20	Claude Sonnet 5	Anthropic	71.5
21	GPT-5.6 Luna	OpenAI	71.5
22	Muse Spark 1.1	Meta	71.3
23	GPT-5.4	OpenAI	71.0
24	Gemini 3.5 Flash	Google	70.1
25	Gemini 3.6 Flash	Google	69.2
26	DeepSeek V4 Flash 0731	DeepSeek	69.1
27	DeepSeek V4 Pro 0813	DeepSeek	68.8
27	Gemini 3.1 Pro	Google	68.8
29	GLM-5.2	Z.AI	68.8
30	Qwen3.8-27B	Alibaba	68.1
31	Qwen3.7 Max	Alibaba	66.0
32	Kimi K2.6	Moonshot AI	61.8
33	MCQuasar 438B	Multiverse Computing	61.2
34	APApodex 1.1	Apodex	60.8
34	APApodex 1.1 Mini	Apodex	60.8
34	Kimi K2.7 Code	Moonshot AI	60.8
37	MiMo-V2.5-Pro	Xiaomi	60.2
38	Hy3	Tencent	58.8
38	Hy3 Preview	Tencent	58.8
40	Muse Spark	Meta	58.6
41	MiniMax M3	MiniMax	58.6
42	GPT-5.4 mini	OpenAI	56.1
43	GPT-5.4 nano	OpenAI	56.1
44	Qwen3.7 Plus	Alibaba	55.9
45	GLM-5.1	Z.AI	55.8
46	Qwen3.6 Plus	Alibaba	54.5
47	Qwen3.6-27B	Alibaba	53.7
48	TMInkling-Small	Thinking Machines Lab	53.0
49	MiniMax M2.7	MiniMax	52.6
50	TMInkling	Thinking Machines Lab	52.1
51	Ling 3.0 Flash	InclusionAI	50.6
51	Ling 3.0 Flash FP8	InclusionAI	50.6
53	MiMo-V2-Flash	Xiaomi	49.8
54	GPT-5.1	OpenAI	49.4
55	Gemini 3.5 Flash-Lite	Google	49.3
56	Nemotron 3 Ultra	NVIDIA	49.3
57	Muse Glimmer 30B	Meta	49.0
58	Mistral Medium 3.5 128B	Mistral	46.9
59	Kimi K2.5	Moonshot AI	46.8
59	Kimi K2.5 (Reasoning)	Moonshot AI	46.8
61	Qwen3.5-122B-A10B	Alibaba	45.7
62	GLM-4.7	Z.AI	45.3
63	Gemma 4 31B	Google	43.4
64	Grok 4.3	xAI	42.3
65	Qwen3.6-35B-A3B	Alibaba	41.9
66	o1	OpenAI	39.7
67	Step 3.7 Flash	StepFun	39.6
68	Gemma 4 26B A4B	Google	39.3
69	GPT-5 (high)	OpenAI	37.8
70	Nemotron 3 Super 100B	NVIDIA	37.7
71	o1-preview	OpenAI	34.0
72	Gemini 2.5 Pro	Google	33.3
73	K-Exaone	LG AI Research	32.1
74	Gemma 4 12B	Google	31.0
75	GPT-OSS 120B	OpenAI	30.4
76	Command A+	Cohere	27.9
77	Nemotron 3.5 Lightning 30B A3B NVFP4	NVIDIA	26.8
78	Mistral Small 4	Mistral	26.6
78	Mistral Small 4 (Reasoning)	Mistral	26.6
80	Trinity-Large-Preview	Arcee AI	25.8
80	Trinity-Large-Thinking	Arcee AI	25.8
82	Ling 2.6 Flash	InclusionAI	25.3
83	Gemini 1.5 Pro	Google	23.6
84	DeepSeek V3	DeepSeek	23.0
85	Granite 4.2 8B	IBM	22.4
86	GPT-4 Turbo	OpenAI	21.5
87	GPT-OSS 20B	OpenAI	20.7
88	GPT-4.1 mini	OpenAI	20.2
89	Mistral Large 3	Mistral	20.1
90	Claude 3 Opus	Anthropic	19.5
91	Llama 4 Maverick	Meta	16.3
92	CECeleris-1	Celeris	14.4
93	Nemotron 3 Nano 30B	NVIDIA	14.4
94	Nemotron 3 Nano Omni 30B A3B	NVIDIA	13.8
95	FAUltravox v0.6 Llama 3.3 70B	Fixie AI	11.9
96	GPT-4o mini	OpenAI	11.4
97	GPT-4.1 nano	OpenAI	11.1
98	Gemma 3 27B	Google	10.1
99	Gemma 4 E4B	Google	9.4
100	Llama 4 Scout	Meta	8.2
101	LFM2.5-2.6B	LiquidAI	7.7
102	Gemma 4 E2B	Google	7.2

Evidence key: Observed

Rows are ordered by the value Artificial Analysis model leaderboards published, highest first. The source does not state whether a higher value is the better result, so this page does not either: for a benchmark that measures a rate of failure, read the table from the bottom.