AA Coding Index

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

이 벤치마크는?

Artificial Analysis가 산출하는 코딩 역량 종합 지수. LiveCodeBench, SWE-bench, Aider 등 실제 코드 생성 및 버그 수정 벤치마크를 종합합니다.

순위	모델
#1	OpenAI GPT-5.4	57.3
#2	Google Gemini 3.1 Pro	55.5
#3	Anthropic Claude Opus 4.7	52.5
#4	Anthropic Claude Opus 4.6	48.1
#5	Anthropic Claude Opus 4.5	47.8
#6	DeepSeek DeepSeek V4 Pro	47.5
#7	Meta Muse Spark	47.5
#8	Z.ai GLM-5	44.2
#9	Z.ai GLM-5.1	43.4
#10	Anthropic Claude Sonnet 4.6	43.0
#11	Alibaba Qwen3.6 Plus	42.9
#12	Google Gemini 3 Flash	42.6
#13	MiniMax MiniMax M2.7	41.9
#14	Xiaomi MiMo-V2-Pro	41.4
#15	Grok Grok 4.20 (Reasoning)	40.5
#16	Moonshot AI Kimi K2.5	39.5
#17	DeepSeek DeepSeek V4 Flash	38.7
#18	Google Gemma 4 31B	38.7
#19	Anthropic Claude Sonnet 4.5	38.6
#20	OpenAI GPT-5.4 Mini	37.5
#21	MiniMax MiniMax M2.5	37.4
#22	Alibaba Qwen3.5 397B A17B	37.4
#23	DeepSeek DeepSeek V3.2	36.7
#24	Anthropic Claude Opus 4.1	36.5
#25	OpenAI GPT-5.4 Nano	35.0
#26	Anthropic Claude Sonnet 4	34.1
#27	Anthropic Claude Opus 4	34.0
#28	Anthropic Claude Haiku 4.5	32.6
#29	Google Gemini 2.5 Pro	31.9
#30	NVIDIA Nemotron 3 Super	31.2
#31	Grok Grok 4.1 Fast (Reasoning)	30.9
#32	Google Gemini 3.1 Flash Lite	30.1
#33	Baidu ERNIE 5.0 Thinking	29.2
#34	Arcee AI Trinity Large Thinking	27.2
#35	Mistral AI Mistral Small 4	24.3
#36	OpenAI GPT-5 Nano	22.9
#37	Google Gemini 2.5 Flash	22.2
#38	Grok Grok 4.20	22.0
#39	OpenAI GPT-5 Mini	21.9
#40	OpenAI GPT-4.1	21.8
#41	OpenAI GPT-5	21.2
#42	Grok Grok 4.1 Fast	19.5
#43	Meituan Longcat Flash Chat	16.5
#44	Meta Llama 4 Maverick	15.6
#45	OpenAI GPT OSS 120B	15.5
#46	Baidu ERNIE 4.5 300B A47B	14.5
#47	Google Gemini 2.5 Flash Lite	14.5
#48	Amazon Nova 2 Lite	12.5
#49	Meta Llama 4 Scout	6.7