모델 순위

순위	변동	모델	점수	95% 구간	투표 수	가격 $/M
1	—	gemini-2.5-pro클로즈드Google	1,456	±5	46,291	—
2	—	claude-opus-4-1-20250805-thinking-16k클로즈드Anthropic	1,449	±6	14,843	—
3	—	o3-2025-04-16클로즈드OpenAI	1,441	±4	46,083	—
4	—	chatgpt-4o-latest-20250326클로즈드OpenAI	1,441	±5	35,807	—
5	—	gpt-4.5-preview-2025-02-27클로즈드OpenAI	1,441	±6	14,644	—
6	—	gpt-5-high클로즈드OpenAI	1,440	±6	17,636	—
7	—	claude-opus-4-1-20250805클로즈드Anthropic	1,438	±6	23,612	—
8	—	gpt-5-chat클로즈드OpenAI	1,430	±6	14,876	—
9	—	claude-opus-4-20250514-thinking-16k클로즈드Anthropic	1,421	±5	30,101	—
10	—	grok-4-0709클로즈드xAI	1,420	±5	23,994	—
11	—	gpt-4.1-2025-04-14클로즈드OpenAI	1,411	±5	36,527	—
12	—	grok-3-preview-02-24클로즈드xAI	1,410	±4	35,154	—
13	—	claude-opus-4-20250514클로즈드Anthropic	1,410	±5	38,104	—
14	—	gemini-2.5-flash클로즈드Google	1,406	±5	47,004	—
15	—	claude-sonnet-4-20250514-thinking-32k클로즈드Anthropic	1,399	±5	28,637	—
16	—	o1-2024-12-17클로즈드OpenAI	1,399	±4	28,039	—
17	—	o4-mini-2025-04-16클로즈드OpenAI	1,392	±5	36,291	—
18	—	mai-1-preview클로즈드Microsoft AI	1,391	±7	9,951	—
19	—	gpt-5-mini-high클로즈드OpenAI	1,389	±7	12,733	—
20	—	claude-sonnet-4-20250514클로즈드Anthropic	1,386	±5	34,172	—
21	—	o1-preview클로즈드OpenAI	1,385	±5	31,505	—
22	—	claude-3-7-sonnet-20250219-thinking-32k클로즈드Anthropic	1,385	±4	36,956	—
23	—	gemini-2.5-flash-lite-preview-06-17-thinking클로즈드Google	1,378	±5	26,274	—
24	—	gpt-4.1-mini-2025-04-14클로즈드OpenAI	1,375	±5	36,061	—
25	—	claude-3-7-sonnet-20250219클로즈드Anthropic	1,369	±4	41,863	—
26	—	claude-3-5-sonnet-20241022클로즈드Anthropic · 변형 1개 더	1,368	±3	86,848	—
27	—	gemma-3-27b-it오픈Google	1,363	±4	39,569	—
28	—	grok-3-mini-high클로즈드xAI	1,362	±6	14,820	—
29	—	gemini-2.0-flash-001클로즈드Google	1,362	±4	42,303	—
30	—	o3-mini-high클로즈드OpenAI	1,361	±5	18,735	—
31	—	grok-3-mini-beta클로즈드xAI	1,357	±5	21,142	—
32	—	gemini-2.0-flash-lite-preview-02-05클로즈드Google	1,351	±4	25,215	—
33	—	gpt-oss-120b오픈OpenAI	1,350	±6	14,753	$0.26
34	—	gemini-1.5-pro-002클로즈드Google	1,350	±3	56,012	—
35	—	o3-mini클로즈드OpenAI	1,346	±4	54,461	—
36	—	llama-3.1-nemotron-ultra-253b-v1오픈NVIDIA	1,344	±12	2,573	—
37	—	gpt-5-nano-high클로즈드OpenAI	1,344	±9	4,775	—
38	—	gpt-4o-2024-05-13클로즈드OpenAI	1,342	±3	113,568	—
39	—	nvidia-llama-3.3-nemotron-super-49b-v1.5오픈NVIDIA	1,340	±10	3,625	—
40	—	gemma-3-12b-it오픈Google	1,339	±10	3,866	—
41	—	o1-mini클로즈드OpenAI	1,333	±4	52,301	—
42	—	llama-3.1-405b-instruct-bf16오픈Meta	1,333	±4	41,932	—
43	—	gpt-4o-2024-08-06클로즈드OpenAI	1,332	±4	45,787	—
44	—	grok-2-2024-08-13클로즈드xAI	1,332	±4	63,725	—
45	—	llama-3.1-405b-instruct-fp8오픈Meta	1,331	±3	60,272	—
46	—	gemini-advanced-0514클로즈드Google	1,331	±5	50,654	—
47	—	gpt-oss-20b오픈OpenAI	1,325	±7	7,197	—
48	—	llama-4-maverick-17b-128e-instruct오픈Meta	1,325	±4	36,588	—
49	—	llama-3.3-nemotron-49b-super-v1오픈NVIDIA	1,324	±12	2,243	—
50	—	gpt-4-turbo-2024-04-09클로즈드OpenAI	1,321	±4	98,965	—
51	—	claude-3-opus-20240229클로즈드Anthropic	1,319	±3	196,368	—
52	—	gemini-1.5-pro-001클로즈드Google	1,319	±4	79,769	—
53	—	llama-4-scout-17b-16e-instruct오픈Meta	1,319	±5	26,364	—
54	—	gpt-4.1-nano-2025-04-14클로즈드OpenAI	1,318	±8	6,143	—
55	—	gemma-3n-e4b-it오픈Google	1,318	±5	24,134	—
56	—	claude-3-5-haiku-20241022클로즈드Anthropic	1,317	±3	66,793	—
57	—	llama-3.3-70b-instruct오픈Meta	1,316	±3	52,593	—
58	—	gpt-4o-mini-2024-07-18클로즈드OpenAI	1,314	±3	69,331	—
59	—	gpt-4-1106-preview클로즈드OpenAI	1,311	±4	101,117	—
60	—	gpt-4-0125-preview클로즈드OpenAI	1,311	±4	94,534	—
61	—	gemini-1.5-flash-002클로즈드Google	1,309	±4	35,180	—
62	—	grok-2-mini-2024-08-13클로즈드xAI	1,305	±4	52,789	—
63	—	gemma-3-4b-it오픈Google	1,301	±9	4,195	—
64	—	llama-3.1-nemotron-70b-instruct오픈NVIDIA	1,296	±8	7,216	—
65	—	llama-3.1-70b-instruct오픈Meta	1,291	±4	56,003	—
66	—	amazon-nova-pro-v1.0클로즈드Amazon	1,286	±4	25,218	—
67	—	llama-3.1-tulu-3-70b오픈Ai2	1,284	±10	2,881	—
68	—	gemma-2-27b-it오픈Google	1,284	±3	76,195	—
69	—	llama-3.1-nemotron-51b-instruct오픈NVIDIA	1,284	±10	3,777	—
70	—	gpt-4-0314클로즈드OpenAI	1,283	±5	54,754	—
71	—	gemini-1.5-flash-001클로즈드Google	1,282	±4	63,418	—
72	—	claude-3-sonnet-20240229클로즈드Anthropic	1,278	±4	110,173	—
73	—	nemotron-4-340b-instruct오픈NVIDIA	1,276	±5	19,913	—
74	—	llama-3-70b-instruct오픈Meta	1,273	±3	158,908	—
75	—	gpt-4-0613클로즈드OpenAI	1,272	±4	89,612	—
76	—	gemma-2-9b-it오픈Google	1,261	±4	54,954	—
77	—	claude-3-haiku-20240307클로즈드Anthropic	1,258	±4	118,626	—
78	—	amazon-nova-lite-v1.0클로즈드Amazon	1,258	±5	19,760	—
79	—	gemini-1.5-flash-8b-001클로즈드Google	1,258	±4	35,914	—
80	—	phi-4오픈Microsoft	1,253	±5	24,354	—
81	—	olmo-2-0325-32b-instruct오픈Ai2	1,251	±11	3,377	—
82	—	amazon-nova-micro-v1.0클로즈드Amazon	1,240	±5	19,774	—
83	—	gemini-pro-dev-api클로즈드Google	1,230	±7	18,454	—
84	—	llama-3-8b-instruct오픈Meta	1,221	±4	106,055	—
85	—	llama-3.1-tulu-3-8b오픈Ai2	1,221	±11	2,943	—
86	—	gpt-3.5-turbo-0125클로즈드OpenAI · 변형 1개 더	1,220	±5	67,214	—
87	—	gemini-pro클로즈드Google	1,217	±12	6,418	—
88	—	llama-3.1-8b-instruct오픈Meta	1,210	±4	50,234	—
89	—	granite-3.1-8b-instruct오픈IBM	1,208	±11	3,142	—
90	—	phi-3-medium-4k-instruct오픈Microsoft	1,196	±5	25,301	—
91	—	gemma-2-2b-it오픈Google	1,195	±4	46,901	—
92	—	dbrx-instruct-preview오픈Databricks	1,193	±6	32,760	—
93	—	wizardlm-70b오픈Microsoft	1,183	±9	8,270	—
94	—	granite-3.0-8b-instruct오픈IBM	1,180	±9	6,727	—
95	—	granite-3.1-2b-instruct오픈IBM	1,179	±11	3,235	—
96	—	snowflake-arctic-instruct오픈Snowflake	1,177	±6	33,272	—
97	—	gemma-1.1-7b-it오픈Google	1,175	±6	24,327	—
98	—	phi-3-small-8k-instruct오픈Microsoft	1,170	±6	17,983	—
99	—	llama-2-70b-chat오픈Meta	1,169	±5	38,767	—
100	—	llama-3.2-3b-instruct오픈Meta	1,165	±8	8,043	—
101	—	llama2-70b-steerlm-chat오픈NVIDIA	1,155	±13	3,605	—
102	—	granite-3.0-2b-instruct오픈IBM	1,154	±8	6,922	—
103	—	wizardlm-13b오픈Microsoft	1,148	±9	7,122	—
104	—	phi-3-mini-4k-instruct-june-2024오픈Microsoft	1,141	±6	12,415	—
105	—	llama-2-13b-chat오픈Meta	1,140	±7	19,357	—
106	—	codellama-34b-instruct오픈Meta	1,134	±9	7,417	—
107	—	palm-2클로즈드Google	1,133	±9	8,634	—
108	—	gemma-7b-it오픈Google	1,131	±9	9,034	—
109	—	phi-3-mini-128k-instruct오픈Microsoft	1,128	±7	21,024	—
110	—	phi-3-mini-4k-instruct오픈Microsoft	1,127	±6	20,539	—
111	—	llama-3.2-1b-instruct오픈Meta	1,111	±8	8,166	—
112	—	gemma-1.1-2b-it오픈Google	1,109	±8	11,035	—
113	—	llama-2-7b-chat오픈Meta	1,107	±7	14,272	—
114	—	gemma-2b-it오픈Google	1,087	±12	4,817	—
115	—	olmo-7b-instruct오픈Ai2	1,073	±11	6,412	—
116	—	dolly-v2-12b오픈Databricks	975	±14	3,471	—
117	—	llama-13b오픈Meta	966	±16	2,441	—

모델 67개 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