Model rankings

Rank	Change	Model	Score	95% interval	Votes	Price $/M
1	—	claude-fable-5ClosedAnthropic	1,508	±9	4,297	—
2	—	claude-opus-4-6-thinkingClosedAnthropic · 1 more variant	1,504	±4	46,410	—
3	—	claude-opus-4-7-thinkingClosedAnthropic · 1 more variant	1,502	±5	32,629	—
4	—	muse-sparkClosedMeta	1,487	±6	13,607	—
5	—	gemini-3.1-pro-previewClosedGoogle	1,486	±4	60,640	—
6	—	gemini-3-proClosedGoogle	1,486	±4	41,314	—
7	—	claude-opus-4-8-thinkingClosedAnthropic · 1 more variant	1,483	±6	12,963	—
8	—	gpt-5.5-highClosedOpenAI	1,481	±5	28,268	—
9	—	gpt-5.4-highClosedOpenAI	1,478	±4	40,959	—
10	—	gemini-3.5-flashClosedGoogle	1,476	±7	10,171	—
11	—	gpt-5.2-chat-latest-20260210ClosedOpenAI	1,475	±4	34,555	—
12	▲ 1	gpt-5.5ClosedOpenAI	1,475	±5	29,071	—
13	▲ 1	grok-4.20-beta-0309-reasoningClosedxAI	1,474	±4	42,370	—
14	▼ 2	grok-4.20-beta1ClosedxAI	1,474	±5	26,964	—
15	—	gemini-3-flashClosedGoogle · 1 more variant	1,473	±4	30,711	—
16	▲ 1	claude-opus-4-5-20251101-thinking-32kClosedAnthropic	1,473	±4	37,087	—
17	▲ 1	gpt-5.5-instantClosedOpenAI	1,473	±5	26,254	—
18	▼ 2	grok-4.20-multi-agent-beta-0309ClosedxAI	1,472	±4	41,384	—
19	—	claude-sonnet-4-6ClosedAnthropic	1,472	±4	39,561	—
20	—	claude-opus-4-5-20251101ClosedAnthropic	1,469	±3	71,167	—
21	—	gpt-5.4ClosedOpenAI	1,468	±4	43,382	—
22	—	grok-4.1-thinkingClosedxAI · 1 more variant	1,466	±3	65,623	—
23	—	claude-sonnet-4-5-20250929-thinking-32kClosedAnthropic	1,455	±3	82,494	—
24	—	claude-sonnet-4-5-20250929ClosedAnthropic	1,455	±3	80,950	—
25	—	gpt-5.1-highClosedOpenAI	1,455	±4	40,820	—
26	—	gemma-4-31bOpenGoogle	1,451	±8	5,884	$0.000
27	—	claude-opus-4-1-20250805-thinking-16kClosedAnthropic	1,449	±3	49,802	—
28	▲ 1	gpt-5.3-chat-latestClosedOpenAI	1,449	±4	33,125	—
29	▼ 1	gpt-5.4-mini-highClosedOpenAI	1,448	±4	39,525	—
30	—	claude-opus-4-1-20250805ClosedAnthropic	1,447	±3	77,333	—
31	—	gemini-2.5-proClosedGoogle	1,446	±3	124,588	—
32	—	gpt-4.5-preview-2025-02-27ClosedOpenAI	1,445	±6	14,547	—
33	—	grok-4.3ClosedxAI	1,444	±5	28,229	—
34	—	chatgpt-4o-latest-20250326ClosedOpenAI	1,443	±3	82,447	—
35	—	gpt-5.1ClosedOpenAI	1,439	±4	43,457	—
36	—	gemma-4-26b-a4bOpenGoogle	1,438	±8	5,813	—
37	—	gpt-5.2-highClosedOpenAI	1,438	±4	48,063	—
38	—	gpt-5.2ClosedOpenAI	1,435	±3	59,625	—
39	—	gpt-5-highClosedOpenAI	1,434	±5	31,928	—
40	—	gemini-3.1-flash-lite-previewClosedGoogle	1,432	±4	48,525	$0.56
41	—	o3-2025-04-16ClosedOpenAI	1,431	±4	59,744	—
42	—	grok-4-1-fast-reasoningClosedxAI	1,431	±3	56,873	—
43	—	amazon-nova-experimental-chat-26-02-10ClosedAmazon	1,427	±10	3,417	—
44	—	gpt-5-chatClosedOpenAI	1,427	±4	31,569	—
45	—	claude-opus-4-20250514-thinking-16kClosedAnthropic	1,424	±4	36,887	—
46	—	grok-4-fast-chatClosedxAI	1,421	±8	6,809	—
47	—	nvidia-nemotron-3-ultra-550b-a55b-nvfp4OpenNVIDIA	1,416	±8	6,153	—
48	—	amazon-nova-experimental-chat-26-01-10ClosedAmazon	1,416	±10	3,406	—
49	—	gpt-4.1-2025-04-14ClosedOpenAI	1,414	±4	50,981	—
50	—	claude-opus-4-20250514ClosedAnthropic	1,412	±4	44,208	—
51	—	grok-3-preview-02-24ClosedxAI	1,412	±4	32,905	—
52	—	claude-haiku-4-5-20251001ClosedAnthropic	1,411	±3	91,153	$2.0
53	—	gemini-2.5-flashClosedGoogle	1,410	±2	124,544	—
54	—	grok-4-0709ClosedxAI	1,410	±4	41,385	—
55	—	gemini-2.5-flash-preview-09-2025ClosedGoogle	1,404	±4	32,910	—
56	—	grok-4-fast-reasoningClosedxAI	1,404	±5	18,710	—
57	—	gpt-5.4-nano-highClosedOpenAI	1,403	±4	38,610	—
58	—	o1-2024-12-17ClosedOpenAI	1,402	±4	27,807	—
59	—	claude-sonnet-4-20250514-thinking-32kClosedAnthropic	1,399	±4	35,108	—
60	—	amazon-nova-experimental-chat-12-10ClosedAmazon	1,395	±10	3,680	—
61	—	gpt-5-mini-highClosedOpenAI	1,390	±5	27,021	—
62	—	o4-mini-2025-04-16ClosedOpenAI	1,390	±4	45,439	—
63	—	claude-sonnet-4-20250514ClosedAnthropic	1,389	±4	40,298	—
64	—	o1-previewClosedOpenAI	1,388	±5	31,122	—
65	—	claude-3-7-sonnet-20250219-thinking-32kClosedAnthropic	1,387	±4	38,819	—
66	—	gpt-4.1-mini-2025-04-14ClosedOpenAI	1,383	±4	39,329	—
67	—	gemini-2.5-flash-lite-preview-09-2025-no-thinkingClosedGoogle	1,380	±3	47,228	—
68	—	gemini-2.5-flash-lite-preview-06-17-thinkingClosedGoogle	1,374	±5	32,899	—
69	—	claude-3-5-sonnet-20241022ClosedAnthropic · 1 more variant	1,373	±3	88,337	—
70	—	claude-3-7-sonnet-20250219ClosedAnthropic	1,371	±4	43,185	—
71	—	amazon-nova-experimental-chat-11-10ClosedAmazon	1,367	±4	25,383	—
72	—	gemma-3-27b-itOpenGoogle	1,366	±4	47,529	—
73	—	o3-mini-highClosedOpenAI	1,363	±5	18,589	—
74	—	grok-3-mini-highClosedxAI	1,362	±5	16,962	—
75	—	nvidia-nemotron-3-super-120b-a12bOpenNVIDIA	1,362	±7	7,544	—
76	—	gemini-2.0-flash-001ClosedGoogle	1,360	±4	43,748	—
77	—	grok-3-mini-betaClosedxAI	1,357	±5	22,715	—
78	—	gemini-2.0-flash-lite-preview-02-05ClosedGoogle	1,353	±4	24,955	—
79	—	gpt-oss-120bOpenOpenAI	1,353	±4	30,635	$0.26
80	—	gemini-1.5-pro-002ClosedGoogle	1,351	±3	55,606	—
81	—	amazon-nova-experimental-chat-10-20ClosedAmazon	1,350	±6	11,470	—
82	▲ 1	amazon-nova-experimental-chat-10-09ClosedAmazon	1,348	±11	2,838	—
83	▼ 1	o3-miniClosedOpenAI	1,348	±4	57,336	—
84	—	llama-3.1-nemotron-ultra-253b-v1OpenNVIDIA	1,347	±12	2,549	—
85	—	mercury-2ClosedInception	1,346	±11	3,124	—
86	—	gpt-4o-2024-05-13ClosedOpenAI	1,346	±3	112,881	—
87	—	nvidia-llama-3.3-nemotron-super-49b-v1.5OpenNVIDIA	1,343	±10	3,345	—
88	—	gemma-3-12b-itOpenGoogle	1,342	±10	3,829	—
89	▲ 1	nova-2-liteClosedAmazon	1,337	±6	12,233	—
90	▼ 1	gpt-5-nano-highClosedOpenAI	1,337	±7	8,266	—
91	—	o1-miniClosedOpenAI	1,337	±4	51,981	—
92	—	grok-2-2024-08-13ClosedxAI	1,336	±4	63,498	—
93	—	gemini-advanced-0514ClosedGoogle	1,335	±5	50,148	—
94	—	gpt-4o-2024-08-06ClosedOpenAI	1,335	±4	45,499	—
95	—	llama-3.1-405b-instruct-bf16OpenMeta	1,335	±4	41,375	—
96	—	llama-3.1-405b-instruct-fp8OpenMeta	1,333	±4	59,656	—
97	—	olmo-3.1-32b-instructOpenAi2	1,330	±6	12,220	—
98	—	llama-3.3-nemotron-49b-super-v1OpenNVIDIA	1,328	±12	2,218	—
99	—	molmo-2-8bOpenAi2	1,327	±21	804	—
100	—	llama-4-maverick-17b-128e-instructOpenMeta	1,327	±4	39,982	—
101	—	gpt-4-turbo-2024-04-09ClosedOpenAI	1,324	±4	98,114	—
102	—	claude-3-5-haiku-20241022ClosedAnthropic	1,324	±3	69,979	—
103	—	gemini-1.5-pro-001ClosedGoogle	1,323	±4	79,138	—
104	—	llama-4-scout-17b-16e-instructOpenMeta	1,323	±5	30,286	—
105	—	gpt-4.1-nano-2025-04-14ClosedOpenAI	1,322	±8	6,103	—
106	—	claude-3-opus-20240229ClosedAnthropic	1,321	±3	194,909	—
107	▲ 1	gemma-3n-e4b-itOpenGoogle	1,318	±5	22,594	—
108	▼ 1	llama-3.3-70b-instructOpenMeta	1,318	±3	54,734	—
109	—	gpt-4o-mini-2024-07-18ClosedOpenAI	1,318	±4	68,709	—
110	—	gpt-oss-20bOpenOpenAI	1,318	±6	10,627	—
111	—	nvidia-nemotron-3-nano-30b-a3b-bf16OpenNVIDIA	1,316	±6	15,506	—
112	—	gpt-4-0125-previewClosedOpenAI	1,313	±4	93,439	—
113	—	gpt-4-1106-previewClosedOpenAI	1,312	±4	100,105	—
114	—	gemini-1.5-flash-002ClosedGoogle	1,309	±4	34,902	—
115	—	grok-2-mini-2024-08-13ClosedxAI	1,308	±4	52,567	—
116	—	granite-4.1-8bOpenIBM	1,307	±10	4,065	—
117	—	mercuryClosedInception	1,306	±14	1,953	—
118	—	olmo-3-32b-thinkOpenAi2	1,305	±8	5,946	—
119	—	gemma-3-4b-itOpenGoogle	1,303	±9	4,171	—
120	—	llama-3.1-nemotron-70b-instructOpenNVIDIA	1,299	±8	7,140	—
121	—	llama-3.1-70b-instructOpenMeta	1,293	±4	55,240	—
122	—	amazon-nova-pro-v1.0ClosedAmazon	1,290	±5	24,745	—
123	—	gemma-2-27b-itOpenGoogle	1,289	±3	75,754	—
124	—	ibm-granite-h-smallOpenIBM	1,287	±8	5,684	—
125	—	gpt-4-0314ClosedOpenAI	1,287	±5	54,173	—
126	—	gemini-1.5-flash-001ClosedGoogle	1,286	±4	62,833	—
127	—	llama-3.1-tulu-3-70bOpenAi2	1,286	±10	2,846	—
128	—	llama-3.1-nemotron-51b-instructOpenNVIDIA	1,286	±10	3,749	—
129	—	olmo-3.1-32b-thinkOpenAi2	1,285	±7	8,501	—
130	—	claude-3-sonnet-20240229ClosedAnthropic	1,280	±4	109,284	—
131	—	nemotron-4-340b-instructOpenNVIDIA	1,276	±5	19,659	—
132	—	llama-3-70b-instructOpenMeta	1,276	±4	156,876	—
133	—	gpt-4-0613ClosedOpenAI	1,275	±4	88,723	—
134	—	gemma-2-9b-itOpenGoogle	1,266	±4	54,611	—
135	—	claude-3-haiku-20240307ClosedAnthropic	1,261	±4	117,701	—
136	—	amazon-nova-lite-v1.0ClosedAmazon	1,260	±5	19,372	—
137	—	gemini-1.5-flash-8b-001ClosedGoogle	1,259	±4	35,558	—
138	—	phi-4OpenMicrosoft	1,256	±5	24,126	—
139	—	olmo-2-0325-32b-instructOpenAi2	1,251	±11	3,334	—
140	—	amazon-nova-micro-v1.0ClosedAmazon	1,241	±5	19,364	—
141	—	gemini-pro-dev-apiClosedGoogle	1,235	±7	18,354	—
142	—	gpt-3.5-turbo-0125ClosedOpenAI · 1 more variant	1,224	±5	66,207	—
143	—	llama-3-8b-instructOpenMeta	1,223	±4	104,642	—
144	—	gemini-proClosedGoogle	1,222	±12	6,390	—
145	—	llama-3.1-tulu-3-8bOpenAi2	1,220	±11	2,896	—
146	—	llama-3.1-8b-instructOpenMeta	1,211	±4	49,605	—
147	—	granite-3.1-8b-instructOpenIBM	1,208	±11	3,090	—
148	—	gemma-2-2b-itOpenGoogle	1,200	±4	46,616	—
149	—	phi-3-medium-4k-instructOpenMicrosoft	1,197	±5	25,055	—
150	—	wizardlm-70bOpenMicrosoft	1,184	±9	8,214	—
151	—	granite-3.0-8b-instructOpenIBM	1,182	±9	6,638	—
152	—	gemma-1.1-7b-itOpenGoogle	1,181	±6	23,893	—
153	—	granite-3.1-2b-instructOpenIBM	1,178	±11	3,188	—
154	—	phi-3-small-8k-instructOpenMicrosoft	1,170	±6	17,766	—
155	—	llama-2-70b-chatOpenMeta	1,170	±6	38,492	—
156	—	llama-3.2-3b-instructOpenMeta	1,166	±8	7,936	—
157	—	granite-3.0-2b-instructOpenIBM	1,156	±8	6,837	—
158	—	llama2-70b-steerlm-chatOpenNVIDIA	1,154	±13	3,585	—
159	—	wizardlm-13bOpenMicrosoft	1,148	±9	7,044	—
160	—	phi-3-mini-4k-instruct-june-2024OpenMicrosoft	1,142	±6	12,297	—
161	—	llama-2-13b-chatOpenMeta	1,141	±7	19,174	—
162	—	palm-2ClosedGoogle	1,137	±9	8,554	—
163	—	gemma-7b-itOpenGoogle	1,137	±9	8,925	—
164	—	codellama-34b-instructOpenMeta	1,136	±9	7,366	—
165	—	phi-3-mini-128k-instructOpenMicrosoft	1,129	±7	20,685	—
166	—	phi-3-mini-4k-instructOpenMicrosoft	1,127	±6	20,118	—
167	—	codellama-70b-instructOpenMeta	1,118	±18	1,143	—
168	—	gemma-1.1-2b-itOpenGoogle	1,115	±8	10,854	—
169	—	llama-3.2-1b-instructOpenMeta	1,110	±8	8,045	—
170	—	llama-2-7b-chatOpenMeta	1,107	±7	14,148	—
171	—	gemma-2b-itOpenGoogle	1,092	±11	4,780	—
172	—	olmo-7b-instructOpenAi2	1,073	±11	6,328	—
173	—	llama-13bOpenMeta	973	±16	2,391	—

123 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