Model rankings

Rank	Change	Model	Score	95% interval	Votes	Price $/M
1	—	gemini-3-proClosedGoogle	1,490	±5	25,178	—
2	—	grok-4.1-thinkingClosedxAI · 1 more variant	1,477	±5	25,440	—
3	—	gemini-3-flashClosedGoogle	1,471	±7	10,204	—
4	—	claude-opus-4-5-20251101-thinking-32kClosedAnthropic	1,469	±6	17,570	—
5	—	claude-opus-4-5-20251101ClosedAnthropic	1,465	±6	18,754	—
6	—	gpt-5.1-highClosedOpenAI	1,457	±5	22,116	—
7	—	gemini-2.5-proClosedGoogle	1,450	±3	84,916	—
8	—	claude-sonnet-4-5-20250929-thinking-32kClosedAnthropic	1,450	±4	36,318	—
9	—	claude-opus-4-1-20250805-thinking-16kClosedAnthropic	1,448	±4	50,339	—
10	—	claude-sonnet-4-5-20250929ClosedAnthropic	1,448	±4	32,372	—
11	—	gpt-4.5-preview-2025-02-27ClosedOpenAI	1,443	±6	14,646	—
12	—	claude-opus-4-1-20250805ClosedAnthropic	1,443	±3	65,088	—
13	—	chatgpt-4o-latest-20250326ClosedOpenAI	1,442	±3	72,793	—
14	—	gpt-5-highClosedOpenAI	1,436	±5	32,723	—
15	—	gpt-5.1ClosedOpenAI	1,435	±5	23,716	—
16	—	gpt-5.2-highClosedOpenAI	1,434	±7	9,657	—
17	—	o3-2025-04-16ClosedOpenAI	1,433	±4	61,321	—
18	—	grok-4-1-fast-reasoningClosedxAI	1,431	±5	18,610	—
19	—	gpt-5.2ClosedOpenAI	1,425	±6	10,415	—
20	—	gpt-5-chatClosedOpenAI	1,425	±4	32,024	—
21	—	claude-opus-4-20250514-thinking-16kClosedAnthropic	1,423	±4	37,671	—
22	—	grok-4-fast-chatClosedxAI	1,420	±8	7,013	—
23	—	gpt-4.1-2025-04-14ClosedOpenAI	1,412	±4	52,315	—
24	—	claude-opus-4-20250514ClosedAnthropic	1,412	±4	45,407	—
25	—	grok-3-preview-02-24ClosedxAI	1,411	±4	34,031	—
26	—	grok-4-0709ClosedxAI	1,409	±4	42,343	—
27	—	gemini-2.5-flashClosedGoogle	1,408	±3	84,420	—
28	—	gemini-2.5-flash-preview-09-2025ClosedGoogle	1,405	±4	33,365	—
29	—	claude-haiku-4-5-20251001ClosedAnthropic	1,403	±4	34,960	$2.0
30	—	grok-4-fast-reasoningClosedxAI	1,402	±5	18,781	—
31	—	o1-2024-12-17ClosedOpenAI	1,401	±4	28,039	—
32	—	claude-sonnet-4-20250514-thinking-32kClosedAnthropic	1,399	±4	36,020	—
33	—	gpt-5-mini-highClosedOpenAI	1,392	±5	27,302	—
34	—	o4-mini-2025-04-16ClosedOpenAI	1,391	±4	46,663	—
35	—	mai-1-previewClosedMicrosoft AI	1,391	±5	18,092	—
36	—	claude-sonnet-4-20250514ClosedAnthropic	1,389	±4	41,446	—
37	—	o1-previewClosedOpenAI	1,388	±5	31,505	—
38	—	claude-3-7-sonnet-20250219-thinking-32kClosedAnthropic	1,387	±4	39,795	—
39	—	gpt-4.1-mini-2025-04-14ClosedOpenAI	1,381	±4	40,335	—
40	—	gemini-2.5-flash-lite-preview-09-2025-no-thinkingClosedGoogle	1,379	±4	37,839	—
41	—	gemini-2.5-flash-lite-preview-06-17-thinkingClosedGoogle	1,375	±4	33,772	—
42	—	claude-3-5-sonnet-20241022ClosedAnthropic · 1 more variant	1,372	±3	89,708	—
43	—	claude-3-7-sonnet-20250219ClosedAnthropic	1,371	±4	44,455	—
44	—	gemma-3-27b-itOpenGoogle	1,365	±4	49,137	—
45	—	o3-mini-highClosedOpenAI	1,363	±5	18,725	—
46	—	grok-3-mini-highClosedxAI	1,362	±5	17,475	—
47	—	amazon-nova-experimental-chat-11-10ClosedAmazon	1,361	±7	7,391	—
48	—	gemini-2.0-flash-001ClosedGoogle	1,360	±4	45,001	—
49	—	grok-3-mini-betaClosedxAI	1,357	±5	23,670	—
50	—	gpt-oss-120bOpenOpenAI	1,353	±4	31,079	$0.26
51	—	gemini-2.0-flash-lite-preview-02-05ClosedGoogle	1,353	±4	25,222	—
52	—	gemini-1.5-pro-002ClosedGoogle	1,351	±3	56,014	—
53	—	o3-miniClosedOpenAI	1,348	±3	58,663	—
54	—	amazon-nova-experimental-chat-10-20ClosedAmazon	1,347	±6	11,615	—
55	—	amazon-nova-experimental-chat-10-09ClosedAmazon	1,347	±11	2,876	—
56	—	llama-3.1-nemotron-ultra-253b-v1OpenNVIDIA	1,346	±12	2,573	—
57	—	gpt-4o-2024-05-13ClosedOpenAI	1,345	±3	113,568	—
58	—	nvidia-llama-3.3-nemotron-super-49b-v1.5OpenNVIDIA	1,341	±10	3,465	—
59	—	gemma-3-12b-itOpenGoogle	1,341	±9	3,866	—
60	—	gpt-5-nano-highClosedOpenAI	1,338	±7	8,348	—
61	—	nova-2-liteClosedAmazon	1,337	±6	12,362	—
62	—	o1-miniClosedOpenAI	1,336	±4	52,303	—
63	—	llama-3.1-405b-instruct-bf16OpenMeta	1,335	±4	41,948	—
64	—	gpt-4o-2024-08-06ClosedOpenAI	1,335	±4	45,787	—
65	—	grok-2-2024-08-13ClosedxAI	1,334	±4	63,725	—
66	—	gemini-advanced-0514ClosedGoogle	1,334	±5	50,652	—
67	—	llama-3.1-405b-instruct-fp8OpenMeta	1,333	±3	60,273	—
68	—	llama-4-maverick-17b-128e-instructOpenMeta	1,327	±4	41,039	—
69	—	llama-3.3-nemotron-49b-super-v1OpenNVIDIA	1,326	±12	2,246	—
70	—	gpt-4-turbo-2024-04-09ClosedOpenAI	1,324	±4	98,965	—
71	—	claude-3-5-haiku-20241022ClosedAnthropic	1,323	±3	71,210	—
72	—	llama-4-scout-17b-16e-instructOpenMeta	1,322	±5	31,044	—
73	—	gemini-1.5-pro-001ClosedGoogle	1,322	±4	79,769	—
74	—	claude-3-opus-20240229ClosedAnthropic	1,322	±3	196,365	—
75	—	gpt-4.1-nano-2025-04-14ClosedOpenAI	1,321	±8	6,146	—
76	—	llama-3.3-70b-instructOpenMeta	1,319	±3	55,942	—
77	—	gemma-3n-e4b-itOpenGoogle	1,319	±5	23,401	—
78	—	gpt-oss-20bOpenOpenAI	1,318	±6	10,820	—
79	—	nvidia-nemotron-3-nano-30b-a3b-bf16OpenNVIDIA	1,317	±7	9,342	—
80	—	gpt-4o-mini-2024-07-18ClosedOpenAI	1,317	±3	69,291	—
81	—	gpt-4-0125-previewClosedOpenAI	1,313	±4	94,534	—
82	—	gpt-4-1106-previewClosedOpenAI	1,313	±4	101,116	—
83	—	mercuryClosedInception	1,312	±14	1,960	—
84	—	gemini-1.5-flash-002ClosedGoogle	1,310	±4	35,180	—
85	—	grok-2-mini-2024-08-13ClosedxAI	1,307	±4	52,792	—
86	—	olmo-3-32b-thinkOpenAi2	1,305	±8	5,977	—
87	—	gemma-3-4b-itOpenGoogle	1,303	±9	4,196	—
88	—	llama-3.1-nemotron-70b-instructOpenNVIDIA	1,298	±8	7,216	—
89	—	llama-3.1-70b-instructOpenMeta	1,293	±4	56,002	—
90	—	amazon-nova-pro-v1.0ClosedAmazon	1,289	±4	25,218	—
91	—	ibm-granite-h-smallOpenIBM	1,288	±8	5,681	—
92	—	gemma-2-27b-itOpenGoogle	1,288	±3	76,196	—
93	—	gpt-4-0314ClosedOpenAI	1,287	±5	54,754	—
94	—	llama-3.1-nemotron-51b-instructOpenNVIDIA	1,286	±10	3,777	—
95	—	llama-3.1-tulu-3-70bOpenAi2	1,286	±10	2,882	—
96	—	gemini-1.5-flash-001ClosedGoogle	1,285	±4	63,418	—
97	—	claude-3-sonnet-20240229ClosedAnthropic	1,281	±4	110,173	—
98	—	nemotron-4-340b-instructOpenNVIDIA	1,277	±5	19,913	—
99	—	llama-3-70b-instructOpenMeta	1,276	±4	158,907	—
100	—	gpt-4-0613ClosedOpenAI	1,275	±4	89,612	—
101	—	gemma-2-9b-itOpenGoogle	1,265	±4	54,957	—
102	—	claude-3-haiku-20240307ClosedAnthropic	1,261	±4	118,626	—
103	—	amazon-nova-lite-v1.0ClosedAmazon	1,260	±5	19,765	—
104	—	gemini-1.5-flash-8b-001ClosedGoogle	1,259	±4	35,914	—
105	—	phi-4OpenMicrosoft	1,256	±5	24,355	—
106	—	olmo-2-0325-32b-instructOpenAi2	1,252	±11	3,379	—
107	—	amazon-nova-micro-v1.0ClosedAmazon	1,241	±5	19,776	—
108	—	gemini-pro-dev-apiClosedGoogle	1,235	±7	18,454	—
109	—	gpt-3.5-turbo-0125ClosedOpenAI · 1 more variant	1,224	±5	67,214	—
110	—	llama-3-8b-instructOpenMeta	1,223	±4	106,055	—
111	—	gemini-proClosedGoogle	1,222	±12	6,418	—
112	—	llama-3.1-tulu-3-8bOpenAi2	1,222	±11	2,943	—
113	—	llama-3.1-8b-instructOpenMeta	1,211	±4	50,235	—
114	—	granite-3.1-8b-instructOpenIBM	1,210	±11	3,142	—
115	—	gemma-2-2b-itOpenGoogle	1,199	±4	46,900	—
116	—	phi-3-medium-4k-instructOpenMicrosoft	1,198	±5	25,301	—
117	—	dbrx-instruct-previewOpenDatabricks	1,196	±6	32,760	—
118	—	wizardlm-70bOpenMicrosoft	1,185	±9	8,270	—
119	—	granite-3.0-8b-instructOpenIBM	1,184	±9	6,727	—
120	—	granite-3.1-2b-instructOpenIBM	1,181	±11	3,236	—
121	—	gemma-1.1-7b-itOpenGoogle	1,180	±6	24,327	—
122	—	snowflake-arctic-instructOpenSnowflake	1,180	±6	33,272	—
123	—	phi-3-small-8k-instructOpenMicrosoft	1,172	±6	17,983	—
124	—	llama-2-70b-chatOpenMeta	1,172	±5	38,767	—
125	—	llama-3.2-3b-instructOpenMeta	1,167	±8	8,043	—
126	—	llama2-70b-steerlm-chatOpenNVIDIA	1,157	±13	3,605	—
127	—	granite-3.0-2b-instructOpenIBM	1,157	±8	6,922	—
128	—	wizardlm-13bOpenMicrosoft	1,150	±9	7,122	—
129	—	phi-3-mini-4k-instruct-june-2024OpenMicrosoft	1,143	±6	12,414	—
130	—	llama-2-13b-chatOpenMeta	1,143	±7	19,357	—
131	—	palm-2ClosedGoogle	1,137	±9	8,634	—
132	—	codellama-34b-instructOpenMeta	1,137	±9	7,417	—
133	—	gemma-7b-itOpenGoogle	1,136	±9	9,034	—
134	—	phi-3-mini-128k-instructOpenMicrosoft	1,131	±7	21,024	—
135	—	phi-3-mini-4k-instructOpenMicrosoft	1,129	±6	20,539	—
136	—	codellama-70b-instructOpenMeta	1,119	±18	1,151	—
137	—	gemma-1.1-2b-itOpenGoogle	1,115	±8	11,035	—
138	—	llama-3.2-1b-instructOpenMeta	1,113	±8	8,166	—
139	—	llama-2-7b-chatOpenMeta	1,109	±7	14,272	—
140	—	gemma-2b-itOpenGoogle	1,092	±12	4,817	—
141	—	olmo-7b-instructOpenAi2	1,074	±11	6,412	—
142	—	dolly-v2-12bOpenDatabricks	980	±14	3,471	—
143	—	llama-13bOpenMeta	972	±16	2,441	—

93 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