Model rankings

Rank	Change	Model	Score	95% interval	Votes	Price $/M
1	—	deepseek-v3-0324OpenDeepSeek · 1 more variant	1,388	±4	15,122	—
2	—	deepseek-r1OpenDeepSeek	1,384	±4	19,430	—
3	—	gemma-3-27b-itOpenGoogle	1,353	±4	18,934	—
4	—	qwen3-235b-a22bOpenAlibaba	1,341	±6	9,327	—
5	—	command-a-03-2025OpenCohere	1,336	±4	17,572	—
6	—	gemma-3-12b-itOpenGoogle	1,330	±9	3,976	—
7	—	llama-3.1-405b-instruct-bf16OpenMeta	1,323	±2	43,788	—
8	—	llama-3.1-405b-instruct-fp8OpenMeta	1,322	±2	63,038	—
9	—	llama-4-maverick-17b-128e-instructOpenMeta	1,319	±4	13,749	—
10	—	llama-3.3-nemotron-49b-super-v1OpenNVIDIA	1,315	±11	2,371	—
11	—	qwq-32bOpenAlibaba · 1 more variant	1,310	±5	15,130	—
12	—	deepseek-v2.5-1210OpenDeepSeek · 1 more variant	1,309	±6	7,243	—
13	—	llama-3.3-70b-instructOpenMeta	1,303	±3	43,904	—
14	—	mistral-large-2407OpenMistral · 1 more variant	1,301	±2	48,217	—
15	—	qwen-max-0919OpenAlibaba	1,300	±4	17,432	—
16	—	athene-v2-chatOpenNexusFlow	1,300	±3	26,074	—
17	—	gemma-3n-e4b-itOpenGoogle	1,298	±8	3,913	—
18	—	athene-70b-0725OpenNexusFlow	1,292	±4	20,580	—
19	—	qwen2.5-72b-instructOpenAlibaba	1,291	±3	41,519	—
20	—	gemma-3-4b-itOpenGoogle	1,289	±7	4,321	—
21	—	llama-3.1-nemotron-70b-instructOpenNVIDIA	1,285	±6	7,577	—
22	—	mistral-small-3.1-24b-instruct-2503OpenMistral	1,285	±11	2,484	—
23	—	llama-3.1-70b-instructOpenMeta	1,283	±2	58,637	—
24	—	llama-3.1-tulu-3-70bOpenAi2	1,278	±9	3,010	—
25	—	gemma-2-27b-itOpenGoogle	1,276	±2	79,538	—
26	—	jamba-1.5-largeOpenAI21 Labs	1,274	±5	9,125	—
27	—	llama-3.1-nemotron-51b-instructOpenNVIDIA	1,273	±8	3,889	—
28	—	gemma-2-9b-it-simpoOpenPrinceton	1,269	±5	10,548	—
29	—	command-r-plus-08-2024OpenCohere	1,266	±5	10,535	—
30	—	nemotron-4-340b-instructOpenNVIDIA	1,265	±3	20,608	—
31	—	llama-3-70b-instructOpenMeta	1,264	±2	163,629	—
32	—	mistral-small-24b-instruct-2501OpenMistral	1,260	±5	15,321	—
33	—	qwen2.5-coder-32b-instructOpenAlibaba	1,259	±6	5,730	—
34	—	c4ai-aya-expanse-32bOpenCohere	1,254	±3	28,768	—
35	—	command-r-plusOpenCohere	1,252	±2	80,846	—
36	—	deepseek-coder-v2OpenDeepSeek	1,251	±4	15,753	—
37	—	qwen2-72b-instructOpenAlibaba	1,249	±3	38,872	—
38	—	gemma-2-9b-itOpenGoogle	1,249	±3	57,197	—
39	—	phi-4OpenMicrosoft	1,246	±3	25,213	—
40	—	command-r-08-2024OpenCohere	1,245	±5	10,851	—
41	—	olmo-2-0325-32b-instructOpenAi2	1,243	±9	3,460	—
42	▲ 1	ministral-8b-2410OpenMistral	1,229	±8	5,111	—
43	▼ 1	jamba-1.5-miniOpenAI21 Labs	1,229	±5	9,274	—
44	—	qwen1.5-110b-chatOpenAlibaba	1,224	±3	27,430	—
45	—	c4ai-aya-expanse-8bOpenCohere	1,220	±4	10,391	—
46	—	mixtral-8x22b-instruct-v0.1OpenMistral	1,218	±2	53,751	—
47	—	qwen1.5-72b-chatOpenAlibaba	1,217	±3	40,658	—
48	—	command-rOpenCohere	1,217	±2	56,398	—
49	—	llama-3.1-tulu-3-8bOpenAi2	1,216	±10	3,074	—
50	—	llama-3-8b-instructOpenMeta	1,211	±2	109,056	—
51	—	granite-3.1-8b-instructOpenIBM	1,204	±9	3,289	—
52	—	yi-1.5-34b-chatOpen01 AI	1,202	±3	25,135	—
53	—	llama-3.1-8b-instructOpenMeta	1,202	±2	52,578	—
54	—	zephyr-orpo-141b-A35b-v0.1OpenHuggingFace	1,199	±9	4,854	—
55	—	qwen1.5-32b-chatOpenAlibaba	1,195	±4	22,765	—
56	—	phi-3-medium-4k-instructOpenMicrosoft	1,187	±3	26,105	—
57	—	gemma-2-2b-itOpenGoogle	1,187	±3	48,892	—
58	—	internlm2_5-20b-chatOpenInternLM	1,185	±5	10,599	—
59	—	mixtral-8x7b-instruct-v0.1OpenMistral	1,183	±2	76,126	—
60	—	dbrx-instruct-previewOpenDatabricks	1,182	±4	33,743	—
61	—	granite-3.0-8b-instructOpenIBM	1,178	±7	7,002	—
62	—	qwen1.5-14b-chatOpenAlibaba	1,176	±4	18,687	—
63	—	wizardlm-70bOpenMicrosoft	1,174	±6	8,383	—
64	—	granite-3.1-2b-instructOpenIBM	1,172	±9	3,380	—
65	—	snowflake-arctic-instructOpenSnowflake	1,171	±3	34,173	—
66	—	yi-34b-chatOpen01 AI	1,170	±4	15,917	—
67	—	openchat-3.5-0106OpenOpenChat · 1 more variant	1,169	±5	12,990	—
68	—	gemma-1.1-7b-itOpenGoogle	1,168	±4	25,070	—
69	—	phi-3-small-8k-instructOpenMicrosoft	1,167	±4	18,476	—
70	—	llama-3.2-3b-instructOpenMeta	1,162	±7	8,390	—
71	—	starling-lm-7b-betaOpenNexusflow	1,161	±4	16,676	—
72	—	tulu-2-dpo-70bOpenAllenAI/UW	1,161	±7	6,658	—
73	—	vicuna-33bOpenLMSYS	1,160	±4	22,936	—
74	—	deepseek-llm-67b-chatOpenDeepSeek	1,159	±8	4,988	—
75	—	openhermes-2.5-mistral-7bOpenNousResearch	1,158	±8	5,088	—
76	—	starling-lm-7b-alphaOpenUC Berkeley	1,152	±5	10,415	—
77	—	llama-2-70b-chatOpenMeta	1,151	±3	39,595	—
78	—	granite-3.0-2b-instructOpenIBM	1,150	±8	7,191	—
79	—	llama2-70b-steerlm-chatOpenNVIDIA	1,142	±9	3,636	—
80	—	nous-hermes-2-mixtral-8x7b-dpoOpenNousResearch	1,140	±8	3,836	—
81	—	dolphin-2.2.1-mistral-7bOpenCognitive Computations	1,138	±13	1,714	—
82	—	phi-3-mini-4k-instruct-june-2024OpenMicrosoft	1,137	±5	12,808	—
83	—	mistral-7b-instruct-v0.2OpenMistral	1,136	±4	20,067	—
84	—	solar-10.7b-instruct-v1.0OpenUpstage AI	1,134	±8	4,286	—
85	—	qwen1.5-7b-chatOpenAlibaba	1,134	±8	4,872	—
86	—	wizardlm-13bOpenMicrosoft	1,132	±7	7,176	—
87	—	mpt-30b-chatOpenMosaicML	1,129	±10	2,644	—
88	—	falcon-180b-chatOpenTII	1,127	±15	1,327	—
89	—	vicuna-13bOpenLMSYS	1,127	±5	19,775	—
90	—	phi-3-mini-4k-instructOpenMicrosoft	1,125	±4	21,097	—
91	—	qwen-14b-chatOpenAlibaba	1,125	±8	5,065	—
92	—	llama-2-13b-chatOpenMeta	1,123	±5	19,722	—
93	—	smollm2-1.7b-instructOpenHuggingFace	1,121	±12	2,375	—
94	—	zephyr-7b-betaOpenHuggingFace	1,118	±5	11,321	—
95	—	phi-3-mini-128k-instructOpenMicrosoft	1,117	±4	21,622	—
96	—	gemma-7b-itOpenGoogle	1,117	±6	9,176	—
97	—	codellama-34b-instructOpenMeta	1,116	±7	7,509	—
98	—	zephyr-7b-alphaOpenHuggingFace	1,115	±13	1,811	—
99	—	llama-3.2-1b-instructOpenMeta	1,114	±7	8,523	—
100	—	guanaco-33bOpenUW	1,109	±11	2,996	—
101	—	gemma-1.1-2b-itOpenGoogle	1,109	±6	11,351	—
102	—	codellama-70b-instructOpenMeta	1,107	±18	1,192	—
103	—	stripedhyena-nous-7bOpenTogether AI	1,102	±8	5,276	—
104	—	mistral-7b-instructOpenMistral	1,095	±6	9,142	—
105	—	llama-2-7b-chatOpenMeta	1,094	±4	14,532	—
106	—	vicuna-7bOpenLMSYS	1,090	±6	7,017	—
107	—	qwen1.5-4b-chatOpenAlibaba	1,083	±6	7,816	—
108	—	gemma-2b-itOpenGoogle	1,077	±8	4,918	—
109	—	olmo-7b-instructOpenAi2	1,063	±8	6,503	—
110	—	koala-13bOpenUC Berkeley	1,048	±8	7,020	—
111	—	chatglm3-6bOpenTsinghua	1,043	±8	4,763	—
112	—	gpt4all-13b-snoozyOpenNomic AI	1,040	±13	1,788	—
113	—	mpt-7b-chatOpenMosaicML	1,036	±10	3,997	—
114	—	alpaca-13bOpenStanford	1,035	±9	5,864	—
115	—	RWKV-4-Raven-14BOpenRWKV	1,017	±8	4,920	—
116	—	chatglm2-6bOpenTsinghua	1,011	±10	2,713	—
117	—	oasst-pythia-12bOpenOpenAssistant	999	±8	6,368	—
118	—	fastchat-t5-3bOpenLMSYS	965	±11	4,288	—
119	—	chatglm-6bOpenTsinghua	957	±9	4,983	—
120	—	dolly-v2-12bOpenDatabricks	950	±10	3,480	—
121	—	llama-13bOpenMeta	935	±12	2,446	—
122	—	stablelm-tuned-alpha-7bOpenStability AI	927	±10	3,336	—

72 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