모델 순위

순위	변동	모델	점수	95% 구간	투표 수	가격 $/M
1	—	Gemma 4 31B오픈Google	1,452	±7	6,130	$0.000
2	—	Inkling오픈Thinking Machines	1,443	±4	50,137	$1.8
3	—	Gemma 4 26B A4B오픈Google	1,437	±7	6,075	—
4	—	NVIDIA-Nemotron-3-Ultra-550B-A55B (NVFP4)오픈NVIDIA	1,426	±7	11,399	—
5	—	Muse Glimmer오픈Meta	1,425	±10	3,793	$0.52
6	—	NVIDIA-Nemotron-3.5-Super-VL-120B-A12B (NVFP4)오픈NVIDIA	1,415	±10	4,061	—
7	—	Inkling-Small오픈Thinking Machines	1,405	±5	37,665	$0.52
8	—	Gemma 3 27B IT오픈Google	1,365	±4	46,613	—
9	—	NVIDIA-Nemotron-3-Super-120B-A12B오픈NVIDIA	1,360	±7	7,844	—
10	—	gpt-oss-120b오픈OpenAI	1,352	±4	30,133	$0.26
11	—	NVIDIA-Nemotron-3.5-Lightning-30B-A3B (NVFP4)오픈NVIDIA	1,351	±5	25,129	—
12	—	Llama 3.1 Nemotron Ultra 253B v1오픈NVIDIA	1,348	±12	2,549	—
13	—	Llama 3.3 Nemotron Super 49B v1.5오픈NVIDIA	1,343	±10	3,324	—
14	—	Gemma 3 12B IT오픈Google	1,342	±10	3,829	—
15	—	Granite 4.2 30B오픈IBM	1,340	±11	3,314	—
16	—	Llama 3.1 405B Instruct (BF16)오픈Meta · 변형 1개 더	1,335	±4	41,375	—
17	—	Olmo 3.1 32B Instruct오픈Ai2	1,329	±6	11,738	—
18	—	Llama 3.3 Nemotron Super 49B v1오픈NVIDIA	1,328	±12	2,218	—
19	—	Llama 4 Maverick 17B 128E Instruct오픈Meta	1,327	±4	39,397	—
20	—	Llama 4 Scout 17B 16E Instruct오픈Meta	1,322	±5	29,777	—
21	—	Molmo 2 8B오픈Ai2	1,321	±21	820	—
22	—	Llama 3.3 70B Instruct오픈Meta	1,318	±4	54,482	—
23	—	gpt-oss-20b오픈OpenAI	1,318	±6	10,400	—
24	—	Gemma 3n E4B IT오픈Google	1,317	±5	22,106	—
25	—	NVIDIA-Nemotron-3-Nano-30B-A3B (BF16)오픈NVIDIA	1,314	±5	15,663	—
26	—	Olmo 3 32B Think오픈Ai2	1,306	±8	5,794	—
27	—	Granite 4.1 8B오픈IBM	1,304	±10	4,138	—
28	—	Gemma 3 4B IT오픈Google	1,303	±9	4,171	—
29	—	Llama 3.1 Nemotron 70B Instruct오픈NVIDIA	1,299	±8	7,140	—
30	—	Llama 3.1 70B Instruct오픈Meta	1,293	±4	55,240	—
31	—	Gemma 2 27B IT오픈Google	1,290	±3	75,754	—
32	▲ 1	Granite 4.2 8B오픈IBM	1,290	±11	3,196	$0.11
33	▼ 1	Granite 4.2 3B오픈IBM	1,289	±12	3,185	—
34	—	Granite-4.0-H-Small오픈IBM	1,287	±8	5,614	—
35	—	Llama 3.1 Nemotron 51B Instruct오픈NVIDIA	1,287	±10	3,749	—
36	—	Llama 3.1 Tulu 3 70B오픈Ai2	1,286	±10	2,846	—
37	—	Olmo 3.1 32B Think오픈Ai2	1,286	±7	8,244	—
38	—	Nemotron-4 340B Instruct오픈NVIDIA	1,277	±5	19,659	—
39	—	Llama 3 70B Instruct오픈Meta	1,277	±4	156,876	—
40	—	Gemma 2 9B IT오픈Google	1,267	±4	54,611	—
41	—	Phi-4오픈Microsoft	1,256	±5	24,126	—
42	—	OLMo 2 32B Instruct (0325)오픈Ai2	1,252	±11	3,334	—
43	—	Llama 3 8B Instruct오픈Meta	1,224	±4	104,642	—
44	—	Llama 3.1 Tulu 3 8B오픈Ai2	1,221	±11	2,896	—
45	—	Llama 3.1 8B Instruct오픈Meta	1,211	±4	49,605	—
46	—	Granite 3.1 8B Instruct오픈IBM	1,209	±11	3,090	—
47	—	Gemma 2 2B IT오픈Google	1,200	±4	46,616	—
48	—	Phi-3 Medium 4K Instruct오픈Microsoft	1,198	±5	25,055	—
49	—	WizardLM 70B오픈Microsoft	1,185	±9	8,214	—
50	—	Granite 3.0 8B Instruct오픈IBM	1,183	±9	6,638	—
51	—	Gemma 1.1 7B IT오픈Google	1,183	±6	23,893	—
52	—	Granite 3.1 2B Instruct오픈IBM	1,179	±11	3,188	—
53	—	Phi-3 Small 8K Instruct오픈Microsoft	1,171	±6	17,766	—
54	—	Llama 2 70B Chat오픈Meta	1,171	±5	38,492	—
55	—	Llama 3.2 3B Instruct오픈Meta	1,167	±8	7,936	—
56	—	Granite 3.0 2B Instruct오픈IBM	1,157	±8	6,837	—
57	—	Llama 2 70B SteerLM Chat오픈NVIDIA	1,155	±13	3,585	—
58	—	WizardLM 13B오픈Microsoft	1,150	±9	7,044	—
59	—	Phi-3 Mini 4K Instruct (June 2024)오픈Microsoft · 변형 1개 더	1,143	±6	12,297	—
60	—	Llama 2 13B Chat오픈Meta	1,141	±7	19,174	—
61	—	Gemma 7B IT오픈Google	1,138	±9	8,925	—
62	—	Code Llama 34B Instruct오픈Meta	1,137	±9	7,366	—
63	—	Phi-3 Mini 128K Instruct오픈Microsoft	1,130	±7	20,685	—
64	—	Code Llama 70B Instruct오픈Meta	1,119	±18	1,143	—
65	—	Gemma 1.1 2B IT오픈Google	1,117	±8	10,854	—
66	—	Llama 3.2 1B Instruct오픈Meta	1,111	±8	8,045	—
67	—	Llama 2 7B Chat오픈Meta	1,108	±7	14,148	—
68	—	Gemma 2B IT오픈Google	1,094	±11	4,780	—
69	—	OLMo 7B Instruct오픈Ai2	1,074	±11	6,328	—
70	—	LLaMA 13B오픈Meta	975	±16	2,391	—

모델 20개 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