AA Coding Index

Name: AA Coding Index
Creator: AIB
License: https://www.gnu.org/licenses/gpl-3.0.html

About This Benchmark

A composite coding capability score by Artificial Analysis. Aggregates LiveCodeBench, SWE-bench, Aider, and other real-world coding benchmarks.

Source: Artificial Analysis

Rank	Model
#1	Anthropic Claude Opus 5	78.0
#2	OpenAI GPT-5.6 Sol	77.4
#3	OpenAI GPT-5.6 Terra	76.7
#4	Anthropic Claude Fable 5	76.5
#5	Moonshot AI Kimi K3	76.2
#6	OpenAI GPT-5.5	74.9
#7	Anthropic Claude Opus 4.8	74.3
#8	Anthropic Claude Opus 4.7	73.6
#9	xAI Grok 4.5	72.4
#10	Anthropic Claude Sonnet 5	71.5
#11	OpenAI GPT-5.6 Luna	71.4
#12	Meta Muse Spark 1.1	71.3
#13	OpenAI GPT-5.4	71.1
#14	Google Gemini 3.5 Flash	70.1
#15	Google Gemini 3.6 Flash	69.2
#16	Google Gemini 3.1 Pro	68.8
#17	Z.ai GLM 5.2	68.8
#18	Alibaba Qwen: Qwen3.7 Max	66.0
#19	Moonshot AI Kimi K2.6	61.8
#20	Moonshot AI Kimi K2.7 Code	60.8
#21	Xiaomi MiMo V2.5	60.2
#22	Xiaomi MiMo V2.5 Pro	60.2
#23	DeepSeek DeepSeek V4 Pro	59.4
#24	Tencent Hy3	58.8
#25	MiniMax MiniMax M3	58.6
#26	Meta Muse Spark	58.6
#27	DeepSeek DeepSeek V4 Flash	56.2
#28	Alibaba Qwen3.7 Plus	55.9
#29	Z.ai GLM-5.1	55.8
#30	Alibaba Qwen3.6 Plus	54.5
#31	MiniMax MiniMax M2.7	52.6
#32	Anthropic Claude Sonnet 4.5	52.1
#33	Google Gemini 3.5 Flash-Lite	49.3
#34	NVIDIA Nemotron 3 Ultra	49.3
#35	Anthropic Claude Opus 4.6	48.1
#36	Anthropic Claude Opus 4.5	47.8
#37	Mistral AI Mistral Medium 3.5	46.9
#38	Moonshot AI Kimi K2.5	46.8
#39	Meituan LongCat 2.0	45.3
#40	Alibaba Qwen3.6 Max	44.9
#41	DeepSeek DeepSeek V3.2	44.2
#42	Z.ai GLM-5	44.2
#43	Anthropic Claude Haiku 4.5	43.9
#44	Google Gemma 4 31B	43.4
#45	Anthropic Claude Sonnet 4.6	43.0
#46	Ring AI Ring-2.6-1T	42.8
#47	Google Gemini 3 Flash	42.6
#48	xAI Grok 4.3	42.2
#49	Alibaba Qwen: Qwen3.6 35B A3B	41.9
#50	Xiaomi MiMo V2 Pro	41.4
#51	xAI Grok 4.20 (Reasoning)	40.5
#52	Stepfun Step 3.7 Flash	39.6
#53	NVIDIA Nemotron 3 Super	37.7
#54	Anthropic Claude Sonnet 4	37.6
#55	OpenAI GPT-5.4 Mini	37.5
#56	MiniMax MiniMax M2.5	37.4
#57	Alibaba Qwen3.5 397B A17B	37.4
#58	Anthropic Claude Opus 4.1	36.5
#59	Cohere North Mini Code	36.5
#60	Z.ai GLM 5V Turbo	36.2
#61	OpenAI GPT-5.4 Nano	35.0
#62	Google Gemini 3.1 Flash Lite	34.7
#63	Anthropic Claude Opus 4	34.0
#64	Google Gemini 2.5 Pro	33.3
#65	LG AI Research K-EXAONE	32.1
#66	xAI Grok 4.1 Fast (Reasoning)	30.9
#67	Baidu ERNIE 5.0 Thinking	29.2
#68	Mistral AI Mistral Small 4	26.6
#69	Arcee AI Trinity Large Thinking	25.8
#70	OpenAI GPT-5 Nano	22.9
#71	Google Gemini 2.5 Flash	22.2
#72	xAI Grok 4.20	22.0
#73	OpenAI GPT-5 Mini	21.9
#74	OpenAI GPT-4.1	21.8
#75	OpenAI GPT OSS 120B	21.2
#76	OpenAI GPT-5	21.2
#77	xAI Grok 4.1 Fast	19.5
#78	Meituan Longcat Flash Chat	16.5
#79	Meta Llama 4 Maverick	16.3
#80	Upstage Solar Pro 3	16.2
#81	Google Gemini 2.5 Flash Lite	14.5
#82	NVIDIA Nemotron 3 Nano 30B A3B	14.4
#83	NVIDIA Nemotron 3 Nano Omni	13.8
#84	Amazon Nova 2 Lite	12.5
#85	IBM Granite 4.1 8B	9.5
#86	Meta Llama 4 Scout	8.2