모델 순위

순위	변동	모델	점수	95% 구간	투표 수	가격 $/M
1	—	llama-3.1-405b-instruct-bf16오픈Meta	1,254	±8	8,787	—
2	—	llama-3.1-405b-instruct-fp8오픈Meta	1,251	±4	33,654	—
3	—	mistral-large-2407오픈Mistral	1,228	±4	29,934	—
4	—	qwen2.5-72b-instruct오픈Alibaba	1,219	±7	6,017	—
5	—	athene-70b-0725오픈NexusFlow	1,219	±5	20,690	—
6	—	deepseek-v2.5오픈DeepSeek	1,218	±6	8,362	—
7	—	llama-3.1-70b-instruct오픈Meta	1,211	±4	28,776	—
8	—	gemma-2-27b-it오픈Google	1,204	±4	40,183	—
9	—	jamba-1.5-large오픈AI21 Labs	1,202	±6	9,178	—
10	—	gemma-2-9b-it-simpo오픈Princeton	1,198	±6	10,613	—
11	—	nemotron-4-340b-instruct오픈NVIDIA	1,196	±5	20,639	—
12	—	llama-3-70b-instruct오픈Meta	1,194	±2	163,885	—
13	—	command-r-plus-08-2024오픈Cohere	1,192	±6	7,603	—
14	—	command-r-plus오픈Cohere	1,183	±3	80,872	—
15	—	qwen2-72b-instruct오픈Alibaba	1,181	±4	38,981	—
16	—	deepseek-coder-v2오픈DeepSeek	1,180	±5	15,793	—
17	—	gemma-2-9b-it오픈Google	1,174	±4	31,557	—
18	—	command-r-08-2024오픈Cohere	1,170	±8	7,838	—
19	—	jamba-1.5-mini오픈AI21 Labs	1,156	±6	9,303	—
20	—	qwen1.5-110b-chat오픈Alibaba	1,155	±4	27,451	—
21	—	mixtral-8x22b-instruct-v0.1오픈Mistral	1,149	±3	53,819	—
22	—	qwen1.5-72b-chat오픈Alibaba	1,148	±3	40,607	—
23	—	command-r오픈Cohere	1,148	±3	56,353	—
24	—	llama-3-8b-instruct오픈Meta	1,141	±2	109,319	—
25	—	yi-1.5-34b-chat오픈01 AI	1,132	±4	25,139	—
26	—	zephyr-orpo-141b-A35b-v0.1오픈HuggingFace	1,131	±7	4,861	—
27	—	qwen1.5-32b-chat오픈Alibaba	1,127	±4	22,746	—
28	—	llama-3.1-8b-instruct오픈Meta	1,125	±4	26,825	—
29	—	phi-3-medium-4k-instruct오픈Microsoft	1,118	±4	26,149	—
30	—	mixtral-8x7b-instruct-v0.1오픈Mistral	1,114	±0	76,119	—
31	—	dbrx-instruct-preview오픈Databricks	1,113	±3	33,701	—
32	—	qwen1.5-14b-chat오픈Alibaba	1,107	±4	18,660	—
33	—	gemma-2-2b-it오픈Google	1,106	±5	23,138	—
34	—	wizardlm-70b오픈Microsoft	1,106	±6	8,384	—
35	—	snowflake-arctic-instruct오픈Snowflake	1,102	±3	34,149	—
36	—	openchat-3.5-0106오픈OpenChat · 변형 1개 더	1,101	±5	12,969	—
37	—	yi-34b-chat오픈01 AI	1,100	±4	15,918	—
38	—	phi-3-small-8k-instruct오픈Microsoft	1,098	±4	18,484	—
39	—	gemma-1.1-7b-it오픈Google	1,097	±3	25,069	—
40	—	llama-3.2-3b-instruct오픈Meta	1,097	±9	6,039	—
41	—	tulu-2-dpo-70b오픈AllenAI/UW	1,092	±8	6,662	—
42	—	vicuna-33b오픈LMSYS	1,092	±4	22,928	—
43	—	starling-lm-7b-beta오픈Nexusflow	1,091	±5	16,658	—
44	—	deepseek-llm-67b-chat오픈DeepSeek	1,090	±8	4,978	—
45	—	openhermes-2.5-mistral-7b오픈NousResearch	1,089	±9	5,090	—
46	—	internlm2_5-20b-chat오픈InternLM	1,086	±17	948	—
47	—	starling-lm-7b-alpha오픈UC Berkeley	1,083	±6	10,414	—
48	—	llama-2-70b-chat오픈Meta	1,081	±3	39,612	—
49	—	llama2-70b-steerlm-chat오픈NVIDIA	1,072	±9	3,635	—
50	—	nous-hermes-2-mixtral-8x7b-dpo오픈NousResearch	1,071	±10	3,842	—
51	—	phi-3-mini-4k-instruct-june-2024오픈Microsoft	1,070	±6	12,887	—
52	—	mistral-7b-instruct-v0.2오픈Mistral	1,067	±4	20,053	—
53	—	qwen1.5-7b-chat오픈Alibaba	1,067	±9	4,861	—
54	—	wizardlm-13b오픈Microsoft	1,065	±7	7,187	—
55	—	falcon-180b-chat오픈TII	1,065	±16	1,326	—
56	—	solar-10.7b-instruct-v1.0오픈Upstage AI	1,065	±8	4,288	—
57	—	dolphin-2.2.1-mistral-7b오픈Cognitive Computations	1,064	±14	1,714	—
58	—	mpt-30b-chat오픈MosaicML	1,063	±14	2,649	—
59	—	vicuna-13b오픈LMSYS	1,060	±5	19,786	—
60	—	llama-3.2-1b-instruct오픈Meta	1,058	±10	6,024	—
61	—	qwen-14b-chat오픈Alibaba	1,056	±9	5,065	—
62	—	phi-3-mini-4k-instruct오픈Microsoft	1,055	±5	21,113	—
63	—	llama-2-13b-chat오픈Meta	1,053	±6	19,726	—
64	—	zephyr-7b-beta오픈HuggingFace	1,049	±7	11,322	—
65	—	phi-3-mini-128k-instruct오픈Microsoft	1,049	±5	21,595	—
66	—	gemma-7b-it오픈Google	1,046	±6	9,174	—
67	—	codellama-34b-instruct오픈Meta	1,046	±7	7,505	—
68	—	guanaco-33b오픈UW	1,044	±13	2,997	—
69	—	zephyr-7b-alpha오픈HuggingFace	1,042	±14	1,815	—
70	—	gemma-1.1-2b-it오픈Google	1,037	±7	11,359	—
71	—	codellama-70b-instruct오픈Meta	1,035	±15	1,192	—
72	—	stripedhyena-nous-7b오픈Together AI	1,035	±8	5,267	—
73	—	mistral-7b-instruct오픈Mistral	1,025	±8	9,133	—
74	—	llama-2-7b-chat오픈Meta	1,025	±5	14,546	—
75	—	vicuna-7b오픈LMSYS	1,023	±7	7,016	—
76	—	qwen1.5-4b-chat오픈Alibaba	1,015	±7	7,807	—
77	—	gemma-2b-it오픈Google	1,012	±8	4,914	—
78	—	olmo-7b-instruct오픈Ai2	995	±8	6,491	—
79	—	gpt4all-13b-snoozy오픈Nomic AI	991	±18	1,786	—
80	—	koala-13b오픈UC Berkeley	984	±11	7,033	—
81	—	chatglm3-6b오픈Tsinghua	974	±11	4,763	—
82	—	alpaca-13b오픈Stanford	973	±11	5,869	—
83	—	mpt-7b-chat오픈MosaicML	972	±13	4,015	—
84	—	RWKV-4-Raven-14B오픈RWKV	954	±11	4,932	—
85	—	oasst-pythia-12b오픈OpenAssistant	936	±11	6,378	—
86	—	chatglm2-6b오픈Tsinghua	932	±14	2,708	—
87	—	fastchat-t5-3b오픈LMSYS	904	±12	4,300	—
88	—	chatglm-6b오픈Tsinghua	896	±12	4,994	—
89	—	dolly-v2-12b오픈Databricks	891	±13	3,487	—
90	—	llama-13b오픈Meta	881	±21	2,444	—
91	—	stablelm-tuned-alpha-7b오픈Stability AI	867	±14	3,339	—

모델 41개 더 보기

지능의 발전

각 개발사가 그 시점에 가진 최고 모델의 Artificial Analysis Intelligence Index를 모델 공개 월 기준으로 보여 줍니다.

가장 많이 오름 · Anthropic +57.2현재 최고 · Anthropic 64.6최고와 최저의 차이 18.1점기간 2025년 11월 – 2026년 11월

64.6Anthropic60.0Google56.9OpenAI50.1Meta46.5Kimi