Model rankings

Rank	Change	Model	Intelligence	Price $/M	Speed tok/s	First token
1	—	Qwen3 235B A22B 2507 (Reasoning)OpenAlibaba · 3 more variants	63.6	$2.6	—	—
2	—	gpt-oss-120BOpenOpenAI	61.3	$0.26	—	—
3	—	DeepSeek R1 0528 (May '25)OpenDeepSeek · 1 more variant	58.7	$0.96	—	—
4	—	GLM-4.5OpenZ AI	56.1	$0.96	—	—
5	—	Qwen3 30B A3B 2507 (Reasoning)OpenAlibaba · 3 more variants	53.2	$0.75	—	—
6	—	MiniMax M1 80kOpenMiniMax	53.2	$0.83	—	—
7	—	Llama Nemotron Super 49B v1.5 (Reasoning)OpenNVIDIA · 1 more variant	52.4	$0.000	—	—
8	—	MiniMax M1 40kOpenMiniMax	51.1	$0.83	—	—
9	—	EXAONE 4.0 32B (Reasoning)OpenLG AI Research · 1 more variant	50.7	$0.70	—	—
10	—	GLM-4.5-AirOpenZ AI	49.5	$0.43	—	—
11	—	gpt-oss-20BOpenOpenAI	49.0	$0.088	—	—
12	—	Kimi K2OpenMoonshot AI	48.6	$1.1	—	—
13	—	QwQ 32BOpenAlibaba	47.7	$0.47	—	—
14	—	Llama 3.1 Nemotron Ultra 253B v1 (Reasoning)OpenNVIDIA	46.4	$0.90	—	—
15	▲ 1	Qwen3 Coder 480B A35BOpenAlibaba	45.2	$3.0	—	—
16	▲ 1	Qwen3 32B (Reasoning)OpenAlibaba · 1 more variant	44.3	$2.6	—	—
17	▲ 1	DeepSeek V3 0324 (Mar '25)OpenDeepSeek · 1 more variant	44.0	$0.48	—	—
18	▲ 1	Llama 4 MaverickOpenMeta	41.7	$0.39	—	—
19	▲ 1	DeepSeek R1 0528 Qwen3 8BOpenDeepSeek	41.5	$0.065	—	—
20	▲ 1	DeepSeek R1 Distill Qwen 32BOpenDeepSeek	41.2	$0.20	—	—
21	▲ 1	Qwen3 8B (Reasoning)OpenAlibaba · 1 more variant	40.8	$0.66	—	—
22	▲ 1	Llama 3.3 Nemotron Super 49B v1 (Reasoning)OpenNVIDIA · 1 more variant	40.5	$0.000	—	—
23	▼ 8	Qwen3 14B (Reasoning)OpenAlibaba · 1 more variant	40.3	$1.3	—	—
24	—	DeepSeek R1 Distill Qwen 14BOpenDeepSeek	38.2	$0.88	—	—
25	—	DeepSeek R1 Distill Llama 70BOpenDeepSeek	37.4	$0.77	—	—
26	—	Qwen3 4B (Reasoning)OpenAlibaba · 1 more variant	36.4	$0.40	—	—
27	—	Reka Flash 3OpenReka AI	36.3	$0.35	—	—
28	—	Magistral SmallOpenMistral	36.0	$0.75	—	—
29	—	Llama 3.1 Nemotron Nano 4B v1.1 (Reasoning)OpenNVIDIA	34.1	$0.000	—	—
30	—	Qwen3 Coder 30B A3BOpenAlibaba	33.4	$0.90	—	—
31	—	Llama 4 ScoutOpenMeta	33.1	$0.21	—	—
32	—	Mistral Small 3.2OpenMistral	31.8	$0.15	—	—
33	—	Command AOpenCohere	31.6	$4.4	—	—
34	—	QwQ 32B-PreviewOpenAlibaba	31.5	$0.67	—	—
35	—	Llama 3.3 Instruct 70BOpenMeta	31.2	$0.59	—	—
36	—	Llama 3.1 Instruct 405BOpenMeta	29.3	$3.3	—	—
37	—	Qwen2.5 Instruct 72BOpenAlibaba	29.2	$0.000	—	—
38	—	MiniMax-Text-01OpenMiniMax	29.1	$0.43	—	—
39	—	Llama 3.1 Tulu3 405BOpenAllen Institute for AI	28.7	$0.000	—	—
40	—	Phi-4OpenMicrosoft Azure	28.0	$0.22	—	—
41	—	Mistral Large 2 (Nov '24)OpenMistral · 1 more variant	27.0	$3.0	—	—
42	—	Qwen3 1.7B (Reasoning)OpenAlibaba · 1 more variant	26.9	$0.40	—	—
43	—	Mistral Small 3.1OpenMistral	26.4	$0.15	—	—
44	—	Pixtral LargeOpenMistral	26.1	$3.0	—	—
45	—	Qwen2.5 Instruct 32BOpenAlibaba	26.1	$0.15	—	—
46	—	Llama 3.1 Nemotron Instruct 70BOpenNVIDIA	26.0	$0.16	—	—
47	—	Gemma 3 27B InstructOpenGoogle	25.2	$0.000	—	—
48	—	Qwen2.5 Coder Instruct 32BOpenAlibaba	25.0	$0.15	—	—
49	—	Llama 3.1 Instruct 70BOpenMeta	24.0	$0.76	—	—
50	—	Gemma 3 12B InstructOpenGoogle	24.0	$0.063	—	—
51	—	Mistral Small 3OpenMistral	23.9	$0.15	—	—
52	—	DeepSeek-V2.5 (Dec '24)OpenDeepSeek · 1 more variant	23.9	$0.18	—	—
53	—	Devstral Small (May '25)OpenMistral · 1 more variant	22.7	$0.15	—	—
54	—	DeepSeek R1 Distill Llama 8BOpenDeepSeek	22.6	$0.040	—	—
55	—	R1 1776OpenPerplexity	22.2	$3.5	—	—
56	—	Llama 3.2 Instruct 90B (Vision)OpenMeta	21.9	$0.72	—	—
56	—	Solar MiniOpenUpstage	21.9	$0.15	—	—
58	—	Grok-1OpenxAI	21.2	$0.000	—	—
59	—	Qwen2 Instruct 72BOpenAlibaba	21.1	$0.000	—	—
60	—	Gemma 2 27BOpenGoogle	20.1	$0.80	—	—
61	—	Llama 3.1 Instruct 8BOpenMeta	19.0	$0.078	—	—
62	—	Gemma 3n E4B InstructOpenGoogle	18.5	$0.025	—	—
63	—	DeepHermes 3 - Mistral 24B PreviewOpenNous Research	18.4	$0.000	—	—
64	—	Jamba 1.7 LargeOpenAI21 Labs	17.9	$3.5	—	—
65	—	Jamba 1.5 LargeOpenAI21 Labs	17.7	$3.5	—	—
66	—	Granite 3.3 Instruct 8BOpenIBM	17.6	$0.085	—	—
67	—	Hermes 3 - Llama-3.1 70BOpenNous Research	17.5	$0.15	—	—
68	—	DeepSeek-Coder-V2OpenDeepSeek	17.3	$0.18	—	—
69	—	Jamba 1.6 LargeOpenAI21 Labs	17.1	$3.5	—	—
70	▲ 8	Gemma 3 4B InstructOpenGoogle	16.9	$0.025	—	—
71	▼ 1	Llama 3 Instruct 70BOpenMeta	15.7	$0.75	—	—
72	▼ 1	Mistral Small (Sep '24)OpenMistral	15.7	$0.30	—	—
73	▼ 1	Gemma 3n E4B Instruct Preview (May '25)OpenGoogle	15.2	$0.000	—	—
74	▼ 1	Phi-4 Multimodal InstructOpenMicrosoft Azure	15.1	$0.000	—	—
75	▼ 1	Qwen2.5 Coder Instruct 7BOpenAlibaba	14.9	$0.000	—	—
76	▼ 1	Mixtral 8x22B InstructOpenMistral	14.4	$3.0	—	—
77	▼ 1	Phi-4 Mini InstructOpenMicrosoft Azure	14.2	$0.000	—	—
78	▼ 1	Llama 2 Chat 7BOpenMeta	13.9	$0.10	—	—
79	—	Llama 3.2 Instruct 11B (Vision)OpenMeta	13.2	$0.16	—	—
80	—	Qwen1.5 Chat 110BOpenAlibaba	13.2	$0.000	—	—
81	—	Phi-3 Medium Instruct 14BOpenMicrosoft Azure	12.7	$0.30	—	—
82	—	Pixtral 12B (2409)OpenMistral	11.4	$0.15	—	—
83	—	Qwen3 0.6B (Reasoning)OpenAlibaba · 1 more variant	11.3	$0.40	—	—
84	—	DeepSeek-V2-ChatOpenDeepSeek	11.1	$0.18	—	—
85	—	Gemma 3n E2B InstructOpenGoogle	10.4	$0.000	—	—
86	—	Ministral 8BOpenMistral	10.4	$0.10	—	—
87	—	Gemma 2 9BOpenGoogle	10.2	$0.20	—	—
88	—	Phi-3 Mini Instruct 3.8BOpenMicrosoft Azure	10.1	$0.23	—	—
89	—	Arctic InstructOpenSnowflake	10.0	$0.000	—	—
89	—	Qwen Chat 72BOpenAlibaba	10.0	$1.0	—	—
91	—	Command-R+ (Aug '24)OpenCohere · 3 more variants	9.5	$4.4	—	—
92	—	Llama 3 Instruct 8BOpenMeta	9.5	$0.058	—	—
93	—	DeepSeek Coder V2 Lite InstructOpenDeepSeek	8.4	$0.000	—	—
94	—	Codestral (May '24)OpenMistral	8.4	$0.30	—	—
95	—	Aya Expanse 32BOpenCohere	8.0	$0.75	—	—
96	—	DeepSeek LLM 67B Chat (V1)OpenDeepSeek	7.9	$0.000	—	—
96	—	Llama 2 Chat 70BOpenMeta	7.9	$0.000	—	—
98	—	Llama 2 Chat 13BOpenMeta	7.9	$0.000	—	—
99	—	OpenChat 3.5 (1210)OpenOpenChat	7.7	$0.000	—	—
100	—	DBRX InstructOpenDatabricks	7.7	$0.000	—	—
101	—	Mistral NeMoOpenMistral	7.5	$0.15	—	—
102	—	Llama 3.2 Instruct 3BOpenMeta	7.4	$0.035	—	—
103	—	DeepSeek R1 Distill Qwen 1.5BOpenDeepSeek	7.1	$0.000	—	—
104	—	Jamba 1.5 MiniOpenAI21 Labs	6.3	$0.25	—	—
105	—	Jamba 1.7 MiniOpenAI21 Labs	5.8	$0.25	—	—
106	—	Jamba 1.6 MiniOpenAI21 Labs	5.5	$0.25	—	—
107	—	Mixtral 8x7B InstructOpenMistral	4.8	$0.70	—	—
108	—	DeepHermes 3 - Llama-3.1 8B PreviewOpenNous Research	4.0	$0.000	—	—
109	—	Aya Expanse 8BOpenCohere	3.8	$0.75	—	—
110	—	Qwen Chat 14BOpenAlibaba	1.7	$0.000	—	—
111	—	Codestral-MambaOpenMistral	1.6	$0.000	—	—
112	—	Gemma 3 1B InstructOpenGoogle	1.1	$0.000	—	—
113	—	Llama 3.2 Instruct 1BOpenMeta	1.0	$0.053	—	—
113	—	Llama 65BOpenMeta	1.0	$0.000	—	—
113	—	Mistral 7B InstructOpenMistral	1.0	$0.25	—	—

65 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