Model rankings

Rank	Change	Model	Intelligence	Price $/M	Speed tok/s	First token
1	—	DeepSeek R1 0528 (May '25)OpenDeepSeek · 1 more variant	68.3	$0.96	—	—
2	—	MiniMax M1 80kOpenMiniMax	63.0	$0.83	—	—
3	—	Qwen3 235B A22B (Reasoning)OpenAlibaba · 1 more variant	62.3	$2.6	—	—
4	—	MiniMax M1 40kOpenMiniMax	61.4	$0.83	—	—
5	—	Llama 3.1 Nemotron Ultra 253B v1 (Reasoning)OpenNVIDIA	60.8	$0.90	—	—
6	—	Qwen3 32B (Reasoning)OpenAlibaba · 1 more variant	59.2	$2.6	—	—
7	—	QwQ 32BOpenAlibaba	58.1	$0.33	—	—
8	—	Qwen3 14B (Reasoning)OpenAlibaba · 1 more variant	55.7	$1.3	—	—
9	—	Qwen3 30B A3B (Reasoning)OpenAlibaba · 1 more variant	55.6	$0.75	—	—
10	—	Magistral SmallOpenMistral	54.7	$0.75	—	—
11	—	DeepSeek V3 0324 (Mar '25)OpenDeepSeek · 1 more variant	53.2	$0.48	—	—
12	—	DeepSeek R1 0528 Qwen3 8BOpenDeepSeek	52.2	$0.065	—	—
13	—	DeepSeek R1 Distill Qwen 32BOpenDeepSeek	51.9	$0.30	—	—
14	—	Qwen3 8B (Reasoning)OpenAlibaba · 1 more variant	51.4	$0.66	—	—
15	—	Llama 3.3 Nemotron Super 49B v1 (Reasoning)OpenNVIDIA · 1 more variant	51.2	$0.000	—	—
16	—	Llama 4 MaverickOpenMeta	50.5	$0.39	—	—
17	—	DeepSeek R1 Distill Qwen 14BOpenDeepSeek	49.0	$0.20	—	—
18	—	DeepSeek R1 Distill Llama 70BOpenDeepSeek	48.2	$0.72	—	—
19	—	Qwen3 4B (Reasoning)OpenAlibaba · 1 more variant	47.3	$0.40	—	—
20	—	Reka Flash 3OpenReka AI	47.1	$0.35	—	—
21	—	Llama 3.1 Nemotron Nano 4B v1.1 (Reasoning)OpenNVIDIA	45.0	$0.000	—	—
22	—	Llama 4 ScoutOpenMeta	43.0	$0.24	—	—
23	—	QwQ 32B-PreviewOpenAlibaba	42.6	$0.67	—	—
24	—	Mistral Small 3.2OpenMistral	42.3	$0.15	—	—
25	—	Llama 3.3 Instruct 70BOpenMeta	41.1	$0.62	—	—
26	—	Llama 3.1 Instruct 405BOpenMeta	40.5	$3.3	—	—
27	—	Qwen2.5 Instruct 72BOpenAlibaba	40.4	$0.000	—	—
28	—	MiniMax-Text-01OpenMiniMax	40.2	$0.43	—	—
29	—	Phi-4OpenMicrosoft Azure	40.2	$0.22	—	—
30	—	Command AOpenCohere	40.0	$4.4	—	—
31	—	Llama 3.1 Tulu3 405BOpenAllen Institute for AI	39.9	$0.000	—	—
32	—	Mistral Large 2 (Nov '24)OpenMistral · 1 more variant	38.3	$3.0	—	—
33	—	Qwen3 1.7B (Reasoning)OpenAlibaba · 1 more variant	38.2	$0.40	—	—
34	—	Gemma 3 27B InstructOpenGoogle	37.6	$0.000	—	—
35	—	Pixtral LargeOpenMistral	37.4	$3.0	—	—
36	—	Qwen2.5 Instruct 32BOpenAlibaba	37.4	$0.15	—	—
37	—	Llama 3.1 Nemotron Instruct 70BOpenNVIDIA	37.3	$0.16	—	—
38	—	Qwen2.5 Coder Instruct 32BOpenAlibaba	36.3	$0.15	—	—
39	—	Llama 3.1 Instruct 70BOpenMeta	35.4	$0.76	—	—
40	—	Mistral Small 3.1OpenMistral	35.3	$0.15	—	—
41	—	Mistral Small 3OpenMistral	35.3	$0.15	—	—
42	—	DeepSeek-V2.5 (Dec '24)OpenDeepSeek · 1 more variant	35.3	$0.18	—	—
43	—	DevstralOpenMistral	34.1	$0.15	—	—
44	—	DeepSeek R1 Distill Llama 8BOpenDeepSeek	34.0	$0.040	—	—
45	—	Gemma 3 12B InstructOpenGoogle	33.8	$0.063	—	—
46	—	R1 1776OpenPerplexity	33.7	$3.5	—	—
47	—	Llama 3.2 Instruct 90B (Vision)OpenMeta	33.4	$0.54	—	—
47	—	Solar MiniOpenUpstage	33.4	$0.15	—	—
49	—	Qwen2 Instruct 72BOpenAlibaba	32.6	$0.000	—	—
50	—	Gemma 2 27BOpenGoogle	31.7	$0.80	—	—
51	—	DeepHermes 3 - Mistral 24B PreviewOpenNous Research	30.0	$0.000	—	—
52	—	Jamba 1.5 LargeOpenAI21 Labs	29.3	$3.5	—	—
53	—	Hermes 3 - Llama-3.1 70BOpenNous Research	29.1	$0.000	—	—
54	—	DeepSeek-Coder-V2OpenDeepSeek	29.0	$0.18	—	—
55	—	Jamba 1.6 LargeOpenAI21 Labs	28.8	$3.5	—	—
56	—	Gemma 3n E4B InstructOpenGoogle	28.0	$0.025	—	—
57	—	Llama 3 Instruct 70BOpenMeta	27.5	$0.75	—	—
58	—	Mistral Small (Sep '24)OpenMistral	27.4	$0.30	—	—
59	—	Gemma 3n E4B Instruct Preview (May '25)OpenGoogle	27.0	$0.000	—	—
60	—	Phi-4 Multimodal InstructOpenMicrosoft Azure	26.9	$0.000	—	—
61	—	Qwen2.5 Coder Instruct 7BOpenAlibaba	26.7	$0.000	—	—
62	—	Mixtral 8x22B InstructOpenMistral	26.2	$3.0	—	—
63	—	Phi-4 Mini InstructOpenMicrosoft Azure	26.0	$0.000	—	—
64	—	Gemma 3 4B InstructOpenGoogle	25.4	$0.025	—	—
65	—	Llama 3.2 Instruct 11B (Vision)OpenMeta	25.0	$0.10	—	—
66	—	Qwen1.5 Chat 110BOpenAlibaba	25.0	$0.000	—	—
67	—	Phi-3 Medium Instruct 14BOpenMicrosoft Azure	24.5	$0.30	—	—
68	—	Llama 3.1 Instruct 8BOpenMeta	23.7	$0.10	—	—
69	—	Pixtral 12B (2409)OpenMistral	23.4	$0.15	—	—
70	—	Qwen3 0.6B (Reasoning)OpenAlibaba · 1 more variant	23.2	$0.40	—	—
71	—	DeepSeek-V2-ChatOpenDeepSeek	23.0	$0.18	—	—
72	—	Ministral 8BOpenMistral	22.3	$0.10	—	—
73	—	Gemma 2 9BOpenGoogle	22.2	$0.20	—	—
74	—	Phi-3 Mini Instruct 3.8BOpenMicrosoft Azure	22.1	$0.000	—	—
75	—	Arctic InstructOpenSnowflake	22.0	$0.000	—	—
75	—	Qwen Chat 72BOpenAlibaba	22.0	$1.0	—	—
77	—	Command-R+ (Aug '24)OpenCohere · 3 more variants	21.5	$4.4	—	—
78	—	Llama 3 Instruct 8BOpenMeta	21.5	$0.058	—	—
79	—	DeepSeek Coder V2 Lite InstructOpenDeepSeek	20.5	$0.000	—	—
80	—	Codestral (May '24)OpenMistral	20.4	$0.30	—	—
81	—	Aya Expanse 32BOpenCohere	20.1	$0.75	—	—
82	—	DeepSeek LLM 67B Chat (V1)OpenDeepSeek	20.0	$0.000	—	—
82	—	Llama 2 Chat 70BOpenMeta	20.0	$0.000	—	—
84	—	Llama 2 Chat 13BOpenMeta	19.9	$0.000	—	—
85	—	OpenChat 3.5 (1210)OpenOpenChat	19.8	$0.000	—	—
86	—	DBRX InstructOpenDatabricks	19.8	$0.000	—	—
87	—	Mistral NeMoOpenMistral	19.6	$0.15	—	—
88	—	Llama 3.2 Instruct 3BOpenMeta	19.5	$0.035	—	—
89	—	DeepSeek R1 Distill Qwen 1.5BOpenDeepSeek	19.2	$0.000	—	—
90	—	Jamba 1.5 MiniOpenAI21 Labs	18.4	$0.25	—	—
91	—	Jamba 1.6 MiniOpenAI21 Labs	17.7	$0.25	—	—
92	—	Mixtral 8x7B InstructOpenMistral	17.0	$0.70	—	—
93	—	DeepHermes 3 - Llama-3.1 8B PreviewOpenNous Research	16.2	$0.000	—	—
94	—	Aya Expanse 8BOpenCohere	16.0	$0.75	—	—
95	—	Qwen Chat 14BOpenAlibaba	14.0	$0.000	—	—
96	—	Codestral-MambaOpenMistral	13.9	$0.000	—	—
97	—	Gemma 3 1B InstructOpenGoogle	13.4	$0.000	—	—
98	—	Llama 65BOpenMeta	11.0	$0.000	—	—
99	—	Mistral 7B InstructOpenMistral	10.1	$0.25	—	—
100	—	Llama 3.2 Instruct 1BOpenMeta	9.7	$0.053	—	—
101	—	Llama 2 Chat 7BOpenMeta	8.3	$0.10	—	—

51 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