모델 순위

순위	변동	모델	지능	가격 $/M	속도 tok/s	첫 토큰
1	—	Gemini 3.1 Pro Preview클로즈드Google	57.2	$4.5	129	29초
2	—	GPT-5.4 (xhigh)클로즈드OpenAI	56.8	$5.6	72	206초
3	—	GPT-5.3 Codex (xhigh)클로즈드OpenAI	53.6	$4.8	81	116초
4	—	Claude Opus 4.6 (Adaptive Reasoning, Max Effort)클로즈드Anthropic	53.0	$10	40	17초
5	—	Muse Spark클로즈드Meta	52.2	—	—	—
6	—	Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort)클로즈드Anthropic	51.7	$6.0	48	94초
7	—	GPT-5.2 (xhigh)클로즈드OpenAI	51.3	$4.8	69	173초
8	—	Claude Opus 4.5 (Reasoning)클로즈드Anthropic	49.7	$10	45	14초
9	—	Grok 4.20 0309 v2 (Reasoning)클로즈드xAI	49.3	$3.0	162	13초
10	—	GPT-5.2 Codex (xhigh)클로즈드OpenAI	49.0	$4.8	121	125초
11	—	GPT-5.4 mini (xhigh)클로즈드OpenAI	48.9	$1.7	149	9.8초
12	—	Grok 4.20 0309 (Reasoning)클로즈드xAI	48.5	$3.0	155	17초
13	—	Gemini 3 Pro Preview (high)클로즈드Google	48.4	$4.5	130	33초
14	—	GPT-5.1 (high)클로즈드OpenAI	47.7	$3.4	91	30초
15	—	GPT-5.2 (medium)클로즈드OpenAI	46.6	$4.8	—	—
16	—	Claude Opus 4.6 (Non-reasoning, High Effort)클로즈드Anthropic	46.5	$10	37	1.7초
17	—	Gemini 3 Flash Preview (Reasoning)클로즈드Google	46.4	$1.1	149	7.3초
18	—	GPT-5 (high)클로즈드OpenAI	44.6	$3.4	85	84초
18	—	GPT-5 Codex (high)클로즈드OpenAI	44.6	$3.4	168	10초
20	—	Claude Sonnet 4.6 (Non-reasoning, High Effort)클로즈드Anthropic	44.4	$6.0	44	1.3초
21	—	GPT-5.4 nano (xhigh)클로즈드OpenAI	44.0	$0.46	156	4.4초
22	—	GPT-5.1 Codex (high)클로즈드OpenAI	43.1	$3.4	163	8.2초
23	—	Claude Opus 4.5 (Non-reasoning)클로즈드Anthropic	43.1	$10	45	1.5초
24	—	Claude 4.5 Sonnet (Reasoning)클로즈드Anthropic	43.0	$6.0	40	9.9초
25	—	Claude Sonnet 4.6 (Non-reasoning, Low Effort)클로즈드Anthropic	42.6	$6.0	45	1.3초
26	—	GPT-5 (medium)클로즈드OpenAI	42.0	$3.4	76	44초
27	—	Claude 4.1 Opus (Reasoning)클로즈드Anthropic	42.0	$30	30	9.9초
28	—	Grok 4클로즈드xAI	41.5	$6.0	55	11초
29	—	Gemini 3 Pro Preview (low)클로즈드Google	41.3	$4.5	—	—
30	—	GPT-5 mini (high)클로즈드OpenAI	41.2	$0.69	77	95초
31	—	o3-pro클로즈드OpenAI	40.7	$35	27	128초
32	—	GPT-5 (low)클로즈드OpenAI	39.2	$3.4	71	6.3초
33	—	Gemma 4 31B (Reasoning)오픈Google	39.2	$0.000	36	1.7초
34	—	Claude 4 Opus (Reasoning)클로즈드Anthropic	39.0	$30	30	7.9초
35	—	GPT-5 mini (medium)클로즈드OpenAI	38.9	$0.69	72	16초
36	—	Claude 4 Sonnet (Reasoning)클로즈드Anthropic	38.7	$6.0	43	11초
37	—	GPT-5.1 Codex mini (high)클로즈드OpenAI	38.6	$0.69	200	7.0초
38	—	Grok 4.1 Fast (Reasoning)클로즈드xAI	38.6	$0.28	150	7.6초
39	—	o3클로즈드OpenAI	38.4	$3.5	89	8.4초
40	—	GPT-5.4 nano (medium)클로즈드OpenAI	38.1	$0.46	154	3.8초
41	—	GPT-5.4 mini (medium)클로즈드OpenAI	37.7	$1.7	152	14초
42	—	Claude 4.5 Sonnet (Non-reasoning)클로즈드Anthropic	37.1	$6.0	38	1.4초
43	—	Claude 4.5 Haiku (Reasoning)클로즈드Anthropic	37.1	$2.0	88	10초
44	—	Claude 4.1 Opus (Non-reasoning)클로즈드Anthropic	36.0	$30	29	1.7초
45	—	NVIDIA Nemotron 3 Super 120B A12B (Reasoning)오픈NVIDIA	36.0	$0.41	193	1.1초
46	—	Nova 2.0 Pro Preview (medium)클로즈드Amazon	35.7	$3.4	128	12초
47	—	GPT-5.4 (Non-reasoning)클로즈드OpenAI	35.4	$5.6	54	0.8초
48	—	Grok 4 Fast (Reasoning)클로즈드xAI	35.1	$0.28	201	3.0초
49	—	Gemini 3 Flash Preview (Non-reasoning)클로즈드Google	35.1	$1.1	151	1.2초
50	—	Claude 3.7 Sonnet (Reasoning)클로즈드Anthropic	34.7	$6.0	—	—
51	—	Gemini 2.5 Pro클로즈드Google	34.6	$3.4	118	25초
52	—	Nova 2.0 Lite (high)클로즈드Amazon	34.5	$0.85	145	19초
53	—	GPT-5.2 (Non-reasoning)클로즈드OpenAI	33.6	$4.8	58	0.9초
54	—	Gemini 3.1 Flash-Lite Preview클로즈드Google	33.5	$0.56	236	7.1초
55	—	gpt-oss-120B (high)오픈OpenAI	33.3	$0.26	206	0.9초
56	—	o4-mini (high)클로즈드OpenAI	33.1	$1.9	128	27초
57	—	Claude 4 Opus (Non-reasoning)클로즈드Anthropic	33.0	$30	30	1.5초
57	—	Claude 4 Sonnet (Non-reasoning)클로즈드Anthropic	33.0	$6.0	39	1.1초
59	—	Mercury 2클로즈드Inception	32.8	$0.38	625	4.4초
60	—	Gemma 4 31B (Non-reasoning)오픈Google	32.3	—	—	—
61	—	Grok 3 mini Reasoning (high)클로즈드xAI	32.1	$0.35	190	0.6초
62	—	Nova 2.0 Pro Preview (low)클로즈드Amazon	31.9	$3.4	132	11초
63	—	Gemma 4 26B A4B (Reasoning)오픈Google	31.2	$0.20	—	—
64	—	Gemini 2.5 Flash Preview (Sep '25) (Reasoning)클로즈드Google	31.1	—	—	—
65	—	Claude 4.5 Haiku (Non-reasoning)클로즈드Anthropic	31.1	$2.0	88	0.8초
66	—	Claude 3.7 Sonnet (Non-reasoning)클로즈드Anthropic	30.8	$6.0	—	—
67	—	o1클로즈드OpenAI	30.8	$26	87	40초
68	—	Gemini 2.5 Pro Preview (Mar' 25)클로즈드Google	30.3	—	—	—
69	—	Nova 2.0 Lite (medium)클로즈드Amazon	29.7	$0.85	152	18초
70	—	Grok 4.20 0309 (Non-reasoning)클로즈드xAI	29.7	$3.0	133	0.6초
71	—	Gemini 2.5 Pro Preview (May' 25)클로즈드Google	29.5	$3.4	—	—
72	—	Grok 4.20 0309 v2 (Non-reasoning)클로즈드xAI	29.0	$3.0	149	0.6초
73	—	Grok Code Fast 1클로즈드xAI	28.7	$0.53	140	4.5초
74	▲ 1	Nemotron Cascade 2 30B A3B오픈NVIDIA	28.4	—	—	—
75	▼ 1	Nova 2.0 Omni (medium)클로즈드Amazon	28.0	$0.85	—	—
76	—	GPT-5.1 (Non-reasoning)클로즈드OpenAI	27.4	$3.4	83	1.3초
77	—	Gemma 4 26B A4B (Non-reasoning)오픈Google	27.1	—	—	—
78	—	Gemini 2.5 Flash (Reasoning)클로즈드Google	27.0	$0.85	200	23초
79	—	GPT-5 nano (high)클로즈드OpenAI	26.8	$0.14	135	64초
80	—	GPT-4.1클로즈드OpenAI	26.3	$3.5	84	1.1초
81	—	GPT-5 nano (medium)클로즈드OpenAI	25.9	$0.14	121	48초
82	—	o3-mini클로즈드OpenAI	25.9	$1.9	142	8.4초
83	—	o1-pro클로즈드OpenAI	25.8	$263	—	—
84	—	Gemini 2.5 Flash Preview (Sep '25) (Non-reasoning)클로즈드Google	25.7	—	—	—
85	—	o3-mini (high)클로즈드OpenAI	25.2	$1.9	139	39초
86	—	Grok 3클로즈드xAI	25.2	$6.0	67	1.4초
87	—	Sonar Reasoning Pro클로즈드Perplexity	24.6	—	—	—
88	—	Nova 2.0 Lite (low)클로즈드Amazon	24.6	$0.85	168	9.3초
89	—	gpt-oss-120B (low)오픈OpenAI	24.5	$0.26	196	0.9초
89	—	gpt-oss-20B (high)오픈OpenAI	24.5	$0.10	265	0.7초
91	—	GPT-5.4 nano (Non-Reasoning)클로즈드OpenAI	24.4	$0.46	157	0.6초
92	—	Gemini 2.5 Flash Preview (Reasoning)클로즈드Google	24.3	—	—	—
93	—	NVIDIA Nemotron 3 Nano 30B A3B (Reasoning)오픈NVIDIA	24.3	$0.11	171	2.5초
94	—	GPT-5 (minimal)클로즈드OpenAI	23.9	$3.4	69	1.6초
95	—	o1-preview클로즈드OpenAI	23.7	$29	—	—
96	—	Grok 4.1 Fast (Non-reasoning)클로즈드xAI	23.6	$0.28	148	0.6초
97	—	GPT-5.4 mini (Non-Reasoning)클로즈드OpenAI	23.3	$1.7	141	0.6초
98	—	Nova 2.0 Omni (low)클로즈드Amazon	23.2	$0.85	—	—
99	—	Grok 4 Fast (Non-reasoning)클로즈드xAI	23.1	$0.28	196	0.6초
100	—	Nova 2.0 Pro Preview (Non-reasoning)클로즈드Amazon	23.1	$3.4	120	1.1초
101	—	GPT-4.1 mini클로즈드OpenAI	22.9	$0.70	73	0.9초
102	—	GPT-5 (ChatGPT)클로즈드OpenAI	21.8	$3.4	141	0.9초
103	—	Gemini 2.5 Flash-Lite Preview (Sep '25) (Reasoning)클로즈드Google	21.7	$0.18	—	—
104	—	Grok 3 Reasoning Beta클로즈드xAI	21.6	—	—	—
105	—	gpt-oss-20B (low)오픈OpenAI	20.8	$0.10	244	0.7초
106	—	GPT-5 mini (minimal)클로즈드OpenAI	20.7	$0.69	76	1.2초
107	—	Gemini 2.5 Flash (Non-reasoning)클로즈드Google	20.6	$0.85	176	0.6초
108	—	o1-mini클로즈드OpenAI	20.4	—	—	—
109	—	GPT-4.5 (Preview)클로즈드OpenAI	20.0	—	—	—
110	—	Gemini 2.0 Flash Thinking Experimental (Jan '25)클로즈드Google	19.6	—	—	—
111	—	Gemini 2.5 Flash-Lite Preview (Sep '25) (Non-reasoning)클로즈드Google	19.4	$0.18	—	—
112	—	Nova Premier클로즈드Amazon	19.0	$5.0	31	3.1초
113	—	Gemma 4 E4B (Reasoning)오픈Google	18.8	—	—	—
114	—	Llama Nemotron Super 49B v1.5 (Reasoning)오픈NVIDIA	18.7	$0.18	59	1.5초
115	—	Claude 3.5 Haiku클로즈드Anthropic	18.7	$1.6	—	—
116	—	GPT-4o (Aug '24)클로즈드OpenAI	18.6	$4.4	86	1.1초
117	—	GPT-4o (March 2025, chatgpt-4o-latest)클로즈드OpenAI	18.6	—	—	—
118	—	Gemini 2.0 Flash (Feb '25)클로즈드Google	18.5	$0.26	—	—
119	—	Llama 3.3 Nemotron Super 49B v1 (Reasoning)오픈NVIDIA	18.5	—	—	—
120	—	Llama 4 Maverick오픈Meta	18.4	$0.48	106	1.1초
121	—	Gemini 2.0 Pro Experimental (Feb '25)클로즈드Google	18.1	—	—	—
122	—	Nova 2.0 Lite (Non-reasoning)클로즈드Amazon	18.0	$0.85	153	1.2초
123	—	Claude 3 Opus클로즈드Anthropic	18.0	$30	—	—
124	—	Sonar Reasoning클로즈드Perplexity	17.9	—	—	—
125	—	Gemini 2.5 Flash Preview (Non-reasoning)클로즈드Google	17.8	—	—	—
126	—	Gemini 2.5 Flash-Lite (Reasoning)클로즈드Google	17.6	$0.18	276	26초
127	—	Llama 3.1 Instruct 405B오픈Meta	17.4	$3.7	29	2.6초
128	—	GPT-4o (Nov '24)클로즈드OpenAI	17.3	$4.4	108	1.1초
129	—	Gemini 2.0 Flash (experimental)클로즈드Google	16.8	$0.000	—	—
130	—	Nova 2.0 Omni (Non-reasoning)클로즈드Amazon	16.6	$0.85	173	1.4초
131	—	Gemini 1.5 Pro (Sep '24)클로즈드Google	16.0	—	—	—
132	—	Claude 3.5 Sonnet (Oct '24)클로즈드Anthropic	15.9	$6.0	—	—
133	—	Sonar클로즈드Perplexity	15.5	—	—	—
134	—	Sonar Pro클로즈드Perplexity	15.2	—	—	—
135	—	Gemma 4 E2B (Reasoning)오픈Google	15.2	—	—	—
136	—	Llama 3.1 Nemotron Ultra 253B v1 (Reasoning)오픈NVIDIA	15.0	$0.90	42	2.5초
137	—	NVIDIA Nemotron Nano 12B v2 VL (Reasoning)오픈NVIDIA	14.9	$0.30	133	0.9초
138	—	Gemma 4 E4B (Non-reasoning)오픈Google	14.8	—	—	—
139	—	NVIDIA Nemotron Nano 9B V2 (Reasoning)오픈NVIDIA	14.8	$0.070	132	0.7초
140	—	Gemini 2.0 Flash-Lite (Feb '25)클로즈드Google	14.7	—	—	—
141	—	NVIDIA Nemotron 3 Nano 4B오픈NVIDIA	14.7	—	—	—
142	—	Llama Nemotron Super 49B v1.5 (Non-reasoning)오픈NVIDIA	14.6	$0.18	61	1.5초
143	—	GPT-4o (May '24)클로즈드OpenAI	14.5	$7.5	91	1.4초
144	—	Llama 3.3 Instruct 70B오픈Meta	14.5	$0.62	87	1.4초
145	—	Gemini 2.0 Flash-Lite (Preview)클로즈드Google	14.5	—	—	—
146	—	Llama 3.1 Nemotron Nano 4B v1.1 (Reasoning)오픈NVIDIA	14.4	—	—	—
147	—	Llama 3.3 Nemotron Super 49B v1 (Non-reasoning)오픈NVIDIA	14.3	—	—	—
148	—	Claude 3.5 Sonnet (June '24)클로즈드Anthropic	14.2	$6.0	—	—
149	—	GPT-4o (ChatGPT)클로즈드OpenAI	14.1	—	—	—
150	—	Grok 2 (Dec '24)오픈xAI	13.9	—	—	—
151	—	GPT-5 nano (minimal)클로즈드OpenAI	13.8	$0.14	137	1.0초
152	—	Gemini 1.5 Flash (Sep '24)클로즈드Google	13.8	—	—	—
153	—	GPT-4 Turbo클로즈드OpenAI	13.7	$15	26	2.8초
154	—	Llama 4 Scout오픈Meta	13.5	$0.29	127	0.8초
155	—	Nova Pro클로즈드Amazon	13.5	$1.4	—	—
156	—	Llama 3.1 Nemotron Instruct 70B오픈NVIDIA	13.4	$1.2	37	1.9초
157	—	Grok Beta클로즈드xAI	13.3	—	—	—
158	—	NVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning)오픈NVIDIA	13.2	$0.088	87	0.5초
159	—	NVIDIA Nemotron Nano 9B V2 (Non-reasoning)오픈NVIDIA	13.2	$0.086	158	1.0초
160	—	GPT-4.1 nano클로즈드OpenAI	13.0	$0.18	172	0.7초
161	—	GPT-4클로즈드OpenAI	12.8	$38	—	—
162	—	Gemini 2.5 Flash-Lite (Non-reasoning)클로즈드Google	12.7	$0.18	215	0.5초
163	—	Nova Lite클로즈드Amazon	12.7	$0.11	151	1.1초
164	—	GPT-4o mini클로즈드OpenAI	12.6	$0.26	33	3.6초
165	—	Llama 3.1 Instruct 70B오픈Meta	12.5	$0.56	20	2.1초
166	—	Gemini 2.0 Flash Thinking Experimental (Dec '24)클로즈드Google	12.3	—	—	—
167	—	Claude 3 Haiku클로즈드Anthropic	12.3	$0.50	134	1.2초
168	—	Gemma 4 E2B (Non-reasoning)오픈Google	12.1	—	—	—
169	—	Gemini 1.5 Pro (May '24)클로즈드Google	12.0	—	—	—
170	—	R1 1776오픈Perplexity	12.0	—	—	—
171	—	Llama 3.2 Instruct 90B (Vision)오픈Meta	11.9	$0.72	50	1.0초
172	—	Llama 3.1 Instruct 8B오픈Meta	11.8	$0.10	143	0.9초
173	—	Grok-1오픈xAI	11.7	—	—	—
174	—	Gemini 1.5 Flash-8B클로즈드Google	11.1	—	—	—
175	—	Granite 4.0 H Small오픈IBM	10.8	$0.11	358	10초
176	—	LFM2 24B A2B오픈Liquid AI	10.5	$0.053	93	0.5초
177	—	Gemini 1.5 Flash (May '24)클로즈드Google	10.5	—	—	—
178	—	Phi-4오픈Microsoft Azure	10.4	$0.22	25	2.3초
179	—	Gemma 3 27B Instruct오픈Google	10.3	$0.000	26	2.1초
180	—	Claude 3 Sonnet클로즈드Anthropic	10.3	$6.0	—	—
181	—	Nova Micro클로즈드Amazon	10.3	$0.061	275	0.9초
182	—	Gemini 1.0 Ultra클로즈드Google	10.1	—	—	—
183	—	Phi-3 Mini Instruct 3.8B오픈Microsoft Azure	10.1	—	—	—
184	—	NVIDIA Nemotron Nano 12B v2 VL (Non-reasoning)오픈NVIDIA	10.1	$0.30	164	1.2초
185	—	Gemma 3n E4B Instruct Preview (May '25)오픈Google	10.1	—	—	—
186	—	Phi-4 Multimodal Instruct오픈Microsoft Azure	10.0	$0.000	16	0.8초
187	—	Llama 2 Chat 7B오픈Meta	9.7	$0.10	—	—
188	—	Llama 3.2 Instruct 3B오픈Meta	9.7	$0.085	53	1.0초
189	—	Claude 2.1클로즈드Anthropic	9.3	—	—	—
190	—	Claude 2.0클로즈드Anthropic	9.1	—	—	—
191	—	GPT-3.5 Turbo클로즈드OpenAI	9.0	$0.75	—	—
192	—	Llama 3 Instruct 70B오픈Meta	8.9	$0.87	—	—
193	—	Arctic Instruct오픈Snowflake	8.8	—	—	—
194	—	Gemma 3 12B Instruct오픈Google	8.8	$0.000	27	30초
195	—	LFM 40B클로즈드Liquid AI	8.8	—	—	—
196	—	Llama 3.2 Instruct 11B (Vision)오픈Meta	8.7	$0.25	50	0.7초
197	—	PALM-2클로즈드Google	8.6	—	—	—
198	—	Gemini 1.0 Pro클로즈드Google	8.5	—	—	—
199	—	Phi-4 Mini Instruct오픈Microsoft Azure	8.4	$0.000	42	0.9초
200	—	Llama 2 Chat 70B오픈Meta	8.4	—	—	—
201	—	Llama 2 Chat 13B오픈Meta	8.4	—	—	—
202	—	DBRX Instruct오픈Databricks	8.3	—	—	—
203	—	LFM2.5-1.2B-Thinking오픈Liquid AI	8.1	—	—	—
204	—	LFM2 2.6B오픈Liquid AI	8.0	$0.000	—	—
204	—	LFM2.5-1.2B-Instruct오픈Liquid AI	8.0	$0.000	—	—
206	—	Granite 4.0 H 1B오픈IBM	8.0	—	—	—
207	—	Gemma 3 270M오픈Google	7.7	—	—	—
208	—	Granite 4.0 Micro오픈IBM	7.7	—	—	—
209	—	Claude Instant클로즈드Anthropic	7.4	—	—	—
209	—	Llama 65B오픈Meta	7.4	—	—	—
211	—	Granite 4.0 1B오픈IBM	7.3	—	—	—
212	—	LFM2 8B A1B오픈Liquid AI	7.0	$0.000	—	—
213	—	Granite 3.3 8B (Non-reasoning)오픈IBM	7.0	$0.085	348	23초
214	—	Gemma 3n E4B Instruct오픈Google	6.4	$0.025	15	1.2초
214	—	Llama 3 Instruct 8B오픈Meta	6.4	$0.070	—	—
216	—	LFM2 1.2B오픈Liquid AI	6.3	$0.000	—	—
217	—	Gemma 3 4B Instruct오픈Google	6.3	$0.000	27	1.5초
218	—	Llama 3.2 Instruct 1B오픈Meta	6.3	$0.050	94	1.1초
219	—	LFM2.5-VL-1.6B오픈Liquid AI	6.2	$0.000	—	—
220	—	Granite 4.0 350M오픈IBM	6.1	—	—	—
221	—	Gemma 3 1B Instruct오픈Google	5.6	$0.000	38	0.9초
222	—	Granite 4.0 H 350M오픈IBM	5.4	—	—	—
223	—	Gemma 3n E2B Instruct오픈Google	4.8	$0.000	—	—

모델 173개 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