Model rankings

Rank	Change	Model	Score	95% interval	Votes	Price $/M
1	—	o3-2025-04-16ClosedOpenAI	1,380	±13	2,211	—
2	—	chatgpt-4o-latest-20250326ClosedOpenAI	1,361	±5	9,229	—
3	—	gpt-4.5-preview-2025-02-27ClosedOpenAI	1,357	±4	15,285	—
4	—	grok-3-preview-02-24ClosedxAI	1,327	±5	14,840	—
5	—	gpt-4.1-2025-04-14ClosedOpenAI	1,325	±10	2,927	—
6	—	o4-mini-2025-04-16ClosedOpenAI	1,323	±11	2,070	—
7	—	o1-2024-12-17ClosedOpenAI	1,319	±3	29,044	—
8	—	claude-3-7-sonnet-20250219-thinking-32kClosedAnthropic	1,312	±5	9,952	—
9	—	o1-previewClosedOpenAI	1,303	±3	33,172	—
10	—	claude-3-7-sonnet-20250219ClosedAnthropic	1,298	±5	15,281	—
11	—	gpt-4.1-mini-2025-04-14ClosedOpenAI	1,291	±9	2,700	—
12	—	o3-mini-highClosedOpenAI	1,287	±4	18,934	—
13	—	claude-3-5-sonnet-20241022ClosedAnthropic · 1 more variant	1,286	±2	65,174	—
14	—	gemini-2.0-flash-001ClosedGoogle	1,284	±3	23,060	—
15	—	gemma-3-27b-itOpenGoogle	1,282	±6	10,735	—
16	—	o3-miniClosedOpenAI	1,269	±3	24,941	—
17	—	gemini-2.0-flash-lite-preview-02-05ClosedGoogle	1,269	±3	23,099	—
18	—	gemini-1.5-pro-002ClosedGoogle	1,268	±2	58,652	—
19	—	gpt-4o-2024-05-13ClosedOpenAI	1,264	±2	117,751	—
20	—	o1-miniClosedOpenAI	1,256	±3	54,961	—
21	—	gemini-advanced-0514ClosedGoogle	1,254	±2	52,140	—
22	—	llama-3.1-405b-instruct-bf16OpenMeta	1,254	±2	43,785	—
23	—	llama-3.1-405b-instruct-fp8OpenMeta	1,253	±2	63,056	—
24	—	gpt-4o-2024-08-06ClosedOpenAI	1,250	±2	47,974	—
25	—	llama-4-maverick-17b-128e-instructOpenMeta	1,250	±7	5,772	—
26	—	grok-2-2024-08-13ClosedxAI	1,249	±2	67,096	—
27	—	llama-3.3-nemotron-49b-super-v1OpenNVIDIA	1,247	±11	2,368	—
28	—	gpt-4-turbo-2024-04-09ClosedOpenAI	1,242	±2	102,153	—
29	—	gemini-1.5-pro-001ClosedGoogle	1,241	±2	82,450	—
30	—	gpt-4.1-nano-2025-04-14ClosedOpenAI	1,240	±10	2,787	—
31	—	claude-3-opus-20240229ClosedAnthropic	1,239	±1	202,698	—
32	—	gpt-4o-mini-2024-07-18ClosedOpenAI	1,235	±2	71,378	—
33	—	llama-3.3-70b-instructOpenMeta	1,235	±2	38,099	—
34	—	gpt-4-1106-previewClosedOpenAI	1,235	±2	103,739	—
35	—	claude-3-5-haiku-20241022ClosedAnthropic	1,233	±3	34,282	—
36	—	gpt-4-0125-previewClosedOpenAI	1,231	±2	97,064	—
37	—	gemini-1.5-flash-002ClosedGoogle	1,229	±3	37,024	—
38	—	grok-2-mini-2024-08-13ClosedxAI	1,225	±2	55,446	—
39	—	llama-3.1-nemotron-70b-instructOpenNVIDIA	1,217	±6	7,576	—
40	—	llama-3.1-70b-instructOpenMeta	1,214	±2	58,650	—
41	—	llama-3.1-tulu-3-70bOpenAi2	1,209	±9	3,011	—
42	—	amazon-nova-pro-v1.0ClosedAmazon	1,208	±4	24,768	—
43	—	gemma-2-27b-itOpenGoogle	1,207	±2	79,532	—
44	—	gemini-1.5-flash-001ClosedGoogle	1,204	±2	65,656	—
45	—	llama-3.1-nemotron-51b-instructOpenNVIDIA	1,204	±8	3,889	—
46	—	gpt-4-0314ClosedOpenAI	1,200	±3	55,970	—
47	—	claude-3-sonnet-20240229ClosedAnthropic	1,197	±2	113,053	—
48	—	nemotron-4-340b-instructOpenNVIDIA	1,196	±4	20,614	—
49	—	llama-3-70b-instructOpenMeta	1,195	±2	163,662	—
50	—	gpt-4-0613ClosedOpenAI	1,185	±2	91,638	—
51	—	amazon-nova-lite-v1.0ClosedAmazon	1,182	±4	20,653	—
52	—	gemma-2-9b-itOpenGoogle	1,180	±2	57,206	—
53	—	gemini-1.5-flash-8b-001ClosedGoogle	1,179	±3	37,691	—
54	—	claude-3-haiku-20240307ClosedAnthropic	1,178	±2	122,315	—
55	—	phi-4OpenMicrosoft	1,177	±3	25,228	—
56	—	amazon-nova-micro-v1.0ClosedAmazon	1,164	±4	20,658	—
57	—	gemini-pro-dev-apiClosedGoogle	1,147	±5	18,801	—
58	—	llama-3.1-tulu-3-8bOpenAi2	1,147	±9	3,076	—
59	—	llama-3-8b-instructOpenMeta	1,143	±2	109,091	—
60	—	gpt-3.5-turbo-0125ClosedOpenAI · 1 more variant	1,137	±2	68,869	—
61	—	granite-3.1-8b-instructOpenIBM	1,135	±8	3,293	—
62	—	llama-3.1-8b-instructOpenMeta	1,133	±3	52,586	—
63	—	gemini-proClosedGoogle	1,132	±8	6,556	—
64	—	phi-3-medium-4k-instructOpenMicrosoft	1,118	±4	26,109	—
65	—	gemma-2-2b-itOpenGoogle	1,118	±3	48,902	—
66	—	dbrx-instruct-previewOpenDatabricks	1,113	±3	33,738	—
67	—	granite-3.0-8b-instructOpenIBM	1,110	±8	7,001	—
68	—	wizardlm-70bOpenMicrosoft	1,105	±6	8,385	—
69	—	granite-3.1-2b-instructOpenIBM	1,104	±8	3,383	—
70	—	snowflake-arctic-instructOpenSnowflake	1,102	±3	34,176	—
71	—	gemma-1.1-7b-itOpenGoogle	1,099	±4	25,068	—
72	—	phi-3-small-8k-instructOpenMicrosoft	1,098	±4	18,472	—
73	—	llama-3.2-3b-instructOpenMeta	1,094	±7	8,400	—
74	—	llama-2-70b-chatOpenMeta	1,082	±3	39,605	—
75	—	granite-3.0-2b-instructOpenIBM	1,082	±8	7,186	—
76	—	llama2-70b-steerlm-chatOpenNVIDIA	1,074	±11	3,636	—
77	—	phi-3-mini-4k-instruct-june-2024OpenMicrosoft	1,067	±5	12,804	—
78	—	wizardlm-13bOpenMicrosoft	1,064	±6	7,176	—
79	—	phi-3-mini-4k-instructOpenMicrosoft	1,056	±4	21,094	—
80	—	llama-2-13b-chatOpenMeta	1,054	±3	19,717	—
81	—	phi-3-mini-128k-instructOpenMicrosoft	1,049	±4	21,633	—
82	—	gemma-7b-itOpenGoogle	1,048	±6	9,176	—
83	—	codellama-34b-instructOpenMeta	1,047	±7	7,509	—
84	—	llama-3.2-1b-instructOpenMeta	1,044	±6	8,517	—
85	—	palm-2ClosedGoogle	1,042	±7	8,712	—
86	—	gemma-1.1-2b-itOpenGoogle	1,041	±6	11,348	—
87	—	codellama-70b-instructOpenMeta	1,039	±18	1,190	—
88	—	llama-2-7b-chatOpenMeta	1,025	±4	14,532	—
89	—	gemma-2b-itOpenGoogle	1,008	±8	4,921	—
90	—	olmo-7b-instructOpenAi2	994	±9	6,502	—
91	—	dolly-v2-12bOpenDatabricks	882	±11	3,481	—
92	—	llama-13bOpenMeta	868	±12	2,446	—

42 more models

Intelligence over time

Each maker's best model at the time on the Artificial Analysis Intelligence Index, by the month it was released.

Gained most · Anthropic +57.2Highest now · Anthropic 64.6Gap between highest and lowest 18.1 ptsTime range Nov 2025 – Nov 2026

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