Model rankings

Rank	Change	Model	Score	95% interval	Votes	Price $/M
1	—	chatgpt-4o-latest-20250326ClosedOpenAI	1,362	±9	4,899	—
2	—	gpt-4.5-preview-2025-02-27ClosedOpenAI	1,356	±6	12,312	—
3	—	grok-3-preview-02-24ClosedxAI	1,327	±6	12,391	—
4	—	o1-2024-12-17ClosedOpenAI	1,320	±5	26,722	—
5	—	claude-3-7-sonnet-20250219-thinking-32kClosedAnthropic	1,316	±7	6,675	—
6	—	o1-previewClosedOpenAI	1,303	±4	33,182	—
7	—	claude-3-7-sonnet-20250219ClosedAnthropic	1,301	±6	12,057	—
8	—	o3-mini-highClosedOpenAI	1,286	±5	15,927	—
9	—	claude-3-5-sonnet-20241022ClosedAnthropic · 1 more variant	1,285	±3	63,546	—
10	—	gemini-2.0-flash-001ClosedGoogle	1,283	±5	20,028	—
11	—	gemma-3-27b-itOpenGoogle	1,280	±7	8,420	—
12	—	gemini-1.5-pro-002ClosedGoogle	1,268	±3	58,660	—
13	▲ 1	o3-miniClosedOpenAI	1,268	±4	22,529	—
14	▼ 1	gemini-2.0-flash-lite-preview-02-05ClosedGoogle	1,267	±5	19,889	—
15	—	gpt-4o-2024-05-13ClosedOpenAI	1,263	±3	117,773	—
16	—	o1-miniClosedOpenAI	1,255	±3	54,957	—
17	—	gemini-advanced-0514ClosedGoogle	1,254	±3	52,155	—
18	—	llama-3.1-405b-instruct-bf16OpenMeta	1,253	±3	41,531	—
19	—	llama-3.1-405b-instruct-fp8OpenMeta	1,252	±3	63,065	—
20	—	gpt-4o-2024-08-06ClosedOpenAI	1,250	±3	47,986	—
21	—	grok-2-2024-08-13ClosedxAI	1,249	±3	67,105	—
22	—	llama-3.3-nemotron-49b-super-v1OpenNVIDIA	1,246	±11	2,374	—
23	—	gpt-4-turbo-2024-04-09ClosedOpenAI	1,242	±3	102,172	—
24	—	gemini-1.5-pro-001ClosedGoogle	1,241	±3	82,452	—
25	—	claude-3-opus-20240229ClosedAnthropic	1,239	±2	202,711	—
26	—	llama-3.3-70b-instructOpenMeta	1,235	±4	35,912	—
27	—	gpt-4-1106-previewClosedOpenAI	1,234	±3	103,754	—
28	—	gpt-4o-mini-2024-07-18ClosedOpenAI	1,234	±3	70,321	—
29	—	claude-3-5-haiku-20241022ClosedAnthropic	1,232	±4	30,918	—
30	—	gpt-4-0125-previewClosedOpenAI	1,231	±3	97,097	—
31	—	gemini-1.5-flash-002ClosedGoogle	1,229	±3	37,011	—
32	—	grok-2-mini-2024-08-13ClosedxAI	1,224	±3	55,458	—
33	—	llama-3.1-nemotron-70b-instructOpenNVIDIA	1,216	±7	7,575	—
34	—	llama-3.1-70b-instructOpenMeta	1,213	±3	58,667	—
35	—	llama-3.1-tulu-3-70bOpenAi2	1,208	±9	3,014	—
36	—	amazon-nova-pro-v1.0ClosedAmazon	1,208	±5	23,087	—
37	—	gemma-2-27b-itOpenGoogle	1,207	±3	79,545	—
38	—	gemini-1.5-flash-001ClosedGoogle	1,204	±3	65,674	—
39	—	llama-3.1-nemotron-51b-instructOpenNVIDIA	1,203	±8	3,886	—
40	—	gpt-4-0314ClosedOpenAI	1,200	±3	55,968	—
41	—	claude-3-sonnet-20240229ClosedAnthropic	1,197	±2	113,092	—
42	—	nemotron-4-340b-instructOpenNVIDIA	1,196	±5	20,609	—
43	—	llama-3-70b-instructOpenMeta	1,195	±2	163,676	—
44	—	gpt-4-0613ClosedOpenAI	1,185	±2	91,631	—
45	—	amazon-nova-lite-v1.0ClosedAmazon	1,181	±5	20,757	—
46	—	gemma-2-9b-itOpenGoogle	1,180	±3	57,207	—
47	—	gemini-1.5-flash-8b-001ClosedGoogle	1,179	±3	37,693	—
48	—	claude-3-haiku-20240307ClosedAnthropic	1,177	±3	122,331	—
49	—	phi-4OpenMicrosoft	1,176	±5	23,391	—
50	—	amazon-nova-micro-v1.0ClosedAmazon	1,163	±5	20,748	—
51	—	gemini-pro-dev-apiClosedGoogle	1,147	±5	18,804	—
52	—	llama-3.1-tulu-3-8bOpenAi2	1,147	±11	3,074	—
53	—	llama-3-8b-instructOpenMeta	1,142	±3	109,087	—
54	—	gpt-3.5-turbo-0125ClosedOpenAI · 1 more variant	1,137	±3	68,874	—
55	—	granite-3.1-8b-instructOpenIBM	1,134	±9	3,295	—
56	—	llama-3.1-8b-instructOpenMeta	1,132	±3	52,610	—
57	—	gemini-proClosedGoogle	1,132	±7	6,556	—
58	—	phi-3-medium-4k-instructOpenMicrosoft	1,118	±4	26,113	—
59	—	gemma-2-2b-itOpenGoogle	1,117	±4	48,898	—
60	—	dbrx-instruct-previewOpenDatabricks	1,113	±4	33,747	—
61	—	granite-3.0-8b-instructOpenIBM	1,110	±9	7,000	—
62	—	wizardlm-70bOpenMicrosoft	1,105	±6	8,383	—
63	—	granite-3.1-2b-instructOpenIBM	1,104	±10	3,381	—
64	—	snowflake-arctic-instructOpenSnowflake	1,102	±4	34,178	—
65	—	gemma-1.1-7b-itOpenGoogle	1,099	±4	25,078	—
66	—	phi-3-small-8k-instructOpenMicrosoft	1,098	±4	18,472	—
67	—	llama-3.2-3b-instructOpenMeta	1,093	±6	8,396	—
68	—	granite-3.0-2b-instructOpenIBM	1,082	±6	7,185	—
69	—	llama-2-70b-chatOpenMeta	1,082	±3	39,610	—
70	—	llama2-70b-steerlm-chatOpenNVIDIA	1,074	±10	3,635	—
71	—	phi-3-mini-4k-instruct-june-2024OpenMicrosoft	1,067	±5	12,809	—
72	—	wizardlm-13bOpenMicrosoft	1,064	±7	7,175	—
73	—	phi-3-mini-4k-instructOpenMicrosoft	1,056	±4	21,099	—
74	—	llama-2-13b-chatOpenMeta	1,054	±4	19,726	—
75	—	phi-3-mini-128k-instructOpenMicrosoft	1,049	±4	21,639	—
76	—	gemma-7b-itOpenGoogle	1,048	±6	9,174	—
77	—	codellama-34b-instructOpenMeta	1,047	±7	7,511	—
78	—	llama-3.2-1b-instructOpenMeta	1,044	±6	8,518	—
79	—	palm-2ClosedGoogle	1,042	±6	8,715	—
80	—	gemma-1.1-2b-itOpenGoogle	1,040	±6	11,346	—
81	—	codellama-70b-instructOpenMeta	1,039	±17	1,192	—
82	—	llama-2-7b-chatOpenMeta	1,025	±4	14,533	—
83	—	gemma-2b-itOpenGoogle	1,008	±8	4,920	—
84	—	olmo-7b-instructOpenAi2	994	±6	6,504	—
85	—	dolly-v2-12bOpenDatabricks	883	±10	3,481	—
86	—	llama-13bOpenMeta	869	±12	2,445	—

36 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