Model rankings

Rank	Change	Model	Intelligence	Price $/M	Speed tok/s	First token
1	—	Kimi K3 (max)OpenKimi · 1 more variant	50.2	$6.0	42	3.3 s
2	—	GLM-5.3 (max)OpenZ AI	48.6	$2.2	83	2.1 s
3	—	Qwen3.8 2.4T A95BOpenAlibaba	46.7	$3.0	40	2.8 s
4	—	GLM-5.3-FlashOpenZ AI	46.2	$0.24	51	1.5 s
5	—	Qwen3.8-Flash-NextOpenAlibaba	45.6	$0.23	87	2.6 s
6	—	DeepSeek V4 Pro 0813 (Reasoning, Max Effort)OpenDeepSeek · 3 more variants	42.1	$2.0	71	1.7 s
7	—	GLM-5.2 (max)OpenZ AI · 1 more variant	42.1	$2.2	76	1.6 s
8	—	Qwen3.8 27B (xhigh)OpenAlibaba · 3 more variants	41.4	$1.1	47	4.0 s
9	—	DeepSeek V4 Flash 0731 (Reasoning, Max Effort)OpenDeepSeek · 3 more variants	40.8	$0.66	133	0.9 s
10	▲ 1	Kimi K2.6OpenKimi · 1 more variant	35.8	$1.7	61	2.0 s
11	▼ 1	MiniMax-M3OpenMiniMax	35.7	$0.52	100	1.3 s
12	—	Kimi K2.7 CodeOpenKimi	33.9	$1.7	60	2.9 s
13	—	MiMo-V2.5-ProOpenXiaomi · 1 more variant	32.6	$0.54	35	5.2 s
14	—	Hy3OpenTencent	32.4	$0.24	99	2.7 s
15	▲ 1	GLM-5 (Reasoning)OpenZ AI · 1 more variant	32.4	$1.6	80	1.4 s
16	▼ 1	GLM-5.1 (Reasoning)OpenZ AI · 1 more variant	31.9	$2.0	56	1.7 s
17	—	MiniMax-M2.7OpenMiniMax	30.1	$0.52	69	1.5 s
18	—	MiMo-V2.5OpenXiaomi	29.5	$0.18	52	3.6 s
19	▲ 1	Qwen3.6 27B (Reasoning)OpenAlibaba · 1 more variant	29.0	$1.4	57	3.7 s
20	▲ 1	Kimi K2.5 (Reasoning)OpenKimi · 1 more variant	27.6	$1.1	—	—
21	▼ 2	Ling 3.0 FlashOpenInclusionAI	27.4	$0.11	322	2.5 s
22	—	Qwen3.5 27B (Reasoning)OpenAlibaba · 1 more variant	26.9	$0.83	77	5.7 s
23	—	MiniMax-M2.5OpenMiniMax	26.8	$0.52	97	1.6 s
24	▲ 1	Hy3-preview (Reasoning)OpenTencent · 1 more variant	26.8	$0.100	—	—
25	▲ 2	MiMo-V2-Flash (Feb 2026)OpenXiaomi · 2 more variants	26.4	—	—	—
26	▼ 2	GLM-4.7 (Reasoning)OpenZ AI · 1 more variant	26.2	$1.0	105	1.4 s
27	▲ 5	Qwen3.6 35B A3B (Reasoning)OpenAlibaba · 1 more variant	26.2	$0.84	130	1.9 s
28	▼ 2	Qwen3.5 397B A17B (Reasoning)OpenAlibaba · 1 more variant	26.1	$1.4	90	2.2 s
29	—	Kimi K2 ThinkingOpenKimi	25.9	$1.1	113	1.4 s
30	▼ 2	LongCat 2.0OpenLongCat	25.8	$1.3	51	3.3 s
31	—	DeepSeek V3.2 (Reasoning)OpenDeepSeek · 1 more variant	25.3	$0.32	—	—
32	▼ 2	Qwen3.5 122B A10B (Reasoning)OpenAlibaba · 1 more variant	24.8	$1.1	131	2.3 s
33	—	MiniMax-M2.1OpenMiniMax	24.6	$0.52	92	1.7 s
34	—	Ring-2.6-1TOpenInclusionAI	23.7	$0.85	126	4.0 s
35	—	DeepSeek V3.1 Terminus (Reasoning)OpenDeepSeek · 1 more variant	23.5	$1.9	—	—
36	—	Step 3.7 FlashOpenStepFun	22.9	$0.44	79	2.8 s
37	—	Qwen3.5 35B A3B (Reasoning)OpenAlibaba · 1 more variant	22.6	$0.69	150	2.3 s
38	▲ 1	MiniMax-M2OpenMiniMax	21.7	$0.52	95	1.6 s
39	▼ 1	GLM-4.6 (Reasoning)OpenZ AI · 1 more variant	21.7	$0.96	53	2.6 s
40	—	Ling-2.6-1TOpenInclusionAI	19.6	$0.85	—	—
41	—	Step 3.5 FlashOpenStepFun	19.1	$0.15	228	1.2 s
42	—	DeepSeek V3.2 Exp (Reasoning)OpenDeepSeek · 1 more variant	19.0	$0.32	—	—
43	▲ 1	Kimi K2 0905OpenKimi · 1 more variant	17.2	$1.1	40	1.1 s
44	▲ 1	GLM-4.7-Flash (Reasoning)OpenZ AI · 1 more variant	16.6	$0.15	105	1.0 s
45	▲ 1	DeepSeek V3.2 SpecialeOpenDeepSeek	16.0	—	—	—
46	▼ 3	Ling 3.0 TinyOpenInclusionAI	15.7	$0.000	113	2.2 s
47	▲ 1	DeepSeek V3.1 (Non-reasoning)OpenDeepSeek · 1 more variant	14.8	$0.85	—	—
48	▼ 1	Qwen3.5 9B (Reasoning)OpenAlibaba · 1 more variant	14.8	$0.15	82	1.7 s
49	▲ 1	Qwen3 VL 235B A22B (Reasoning)OpenAlibaba	14.4	$1.3	56	3.0 s
50	▼ 1	Qwen3 Coder NextOpenAlibaba	14.3	$0.56	115	1.2 s
51	—	Qwen3.5 4B (Reasoning)OpenAlibaba · 1 more variant	13.9	$0.060	21	0.8 s
52	—	DeepSeek R1 0528 (May '25)OpenDeepSeek · 1 more variant	13.9	$1.8	—	—
53	▲ 1	GLM-4.5 (Reasoning)OpenZ AI	13.3	—	—	—
54	▼ 1	Qwen3 235B A22B 2507 (Reasoning)OpenAlibaba · 2 more variants	13.1	$0.75	54	2.9 s
55	—	Seed-OSS-36B-InstructOpenByteDance Seed	12.2	$0.30	34	2.7 s
56	—	Qwen3 235B A22B 2507 InstructOpenAlibaba	12.1	$0.40	59	2.4 s
57	—	Qwen3 Coder 480B A35B InstructOpenAlibaba	11.9	$3.0	54	3.0 s
58	—	Qwen3 VL 32B (Reasoning)OpenAlibaba	11.9	$0.28	89	2.7 s
59	—	MiniMax M1 80kOpenMiniMax	11.6	$0.96	—	—
60	—	LongCat Flash LiteOpenLongCat	11.2	—	—	—
61	—	GLM-4.6V (Reasoning)OpenZ AI · 1 more variant	10.7	$0.45	64	4.5 s
62	—	Qwen3 Next 80B A3B (Reasoning)OpenAlibaba	10.7	$0.41	196	2.2 s
63	—	GLM-4.5-AirOpenZ AI	10.5	$0.37	66	2.8 s
64	—	Ring-1TOpenInclusionAI	10.2	—	—	—
65	—	DeepSeek V3 0324OpenDeepSeek · 1 more variant	9.2	$0.48	—	—
66	—	Qwen3 30B A3B 2507 (Reasoning)OpenAlibaba · 2 more variants	8.6	$0.75	159	2.4 s
67	—	MiniMax M1 40kOpenMiniMax	8.5	—	—	—
68	—	Qwen3 VL 235B A22B InstructOpenAlibaba	8.4	$0.70	50	2.6 s
69	—	Ling 2.6 FlashOpenInclusionAI	7.9	—	—	—
70	—	Qwen3 Next 80B A3B InstructOpenAlibaba	7.8	$0.41	179	2.2 s
71	—	Qwen3 Coder 30B A3B InstructOpenAlibaba	7.7	$0.90	91	2.5 s
72	—	QwQ 32BOpenAlibaba	7.5	$0.75	—	—
73	—	Qwen3 VL 30B A3B (Reasoning)OpenAlibaba	7.5	$0.75	110	2.1 s
74	—	Ling-1TOpenInclusionAI	6.9	—	—	—
75	—	Qwen3 4B 2507 (Reasoning)OpenAlibaba · 2 more variants	6.2	—	—	—
76	—	Qwen3 32B (Reasoning)OpenAlibaba · 1 more variant	5.7	$0.28	103	2.5 s
77	—	Qwen3 VL 32B InstructOpenAlibaba	5.3	$0.28	71	2.6 s
78	—	DeepSeek R1 Distill Qwen 32BOpenDeepSeek	5.3	—	—	—
79	—	Qwen3 VL 8B (Reasoning)OpenAlibaba	4.9	$0.66	114	2.3 s
80	—	Qwen3 14B (Reasoning)OpenAlibaba · 1 more variant	4.8	$1.3	59	2.7 s
81	—	DeepSeek R1 0528 Qwen3 8BOpenDeepSeek	4.7	—	—	—
82	—	Qwen3 VL 30B A3B InstructOpenAlibaba	4.3	$0.35	112	2.1 s
83	—	DeepSeek R1 Distill Llama 70BOpenDeepSeek	4.2	$0.80	19	1.3 s
84	—	DeepSeek R1 Distill Qwen 14BOpenDeepSeek	4.1	—	—	—
85	—	Ling-flash-2.0OpenInclusionAI	4.1	$0.25	4	2.6 s
86	—	Qwen3 Omni 30B A3B (Reasoning)OpenAlibaba	4.0	$0.43	100	1.9 s
87	—	Qwen2.5 Instruct 72BOpenAlibaba	3.9	$0.48	—	—
88	—	Step3 VL 10BOpenStepFun	3.8	—	—	—
89	—	QwQ 32B-PreviewOpenAlibaba	3.6	—	—	—
90	—	GLM-4.5V (Reasoning)OpenZ AI · 1 more variant	3.5	$0.90	53	2.5 s
91	—	Qwen3 30B A3B 2507 InstructOpenAlibaba	3.4	$0.35	146	1.8 s
92	—	ERNIE 4.5 300B A47BOpenBaidu	3.4	$0.49	—	—
93	—	Kimi Linear 48B A3B InstructOpenKimi	2.9	—	—	—
94	—	Qwen3 8B (Reasoning)OpenAlibaba · 1 more variant	2.8	$0.66	40	3.7 s
95	—	Qwen3 VL 8B InstructOpenAlibaba	2.8	$0.31	118	2.3 s
96	—	Ring-flash-2.0OpenInclusionAI	2.6	$0.25	—	—
97	—	Qwen3 VL 4B (Reasoning)OpenAlibaba	2.3	—	—	—
98	—	Qwen3.5 2B (Reasoning)OpenAlibaba · 1 more variant	2.1	—	—	—
99	—	Qwen2.5 Instruct 32BOpenAlibaba	1.9	—	—	—
100	—	Qwen3 4B 2507 InstructOpenAlibaba	1.6	—	—	—
101	—	Qwen2.5 Coder Instruct 32BOpenAlibaba	1.6	—	—	—
102	—	DeepSeek-V2.5 (Dec '24)OpenDeepSeek · 1 more variant	1.2	—	—	—
103	▲ 12	DeepSeek Coder V2 Lite InstructOpenDeepSeek	1.0	—	—	—
103	▲ 13	DeepSeek LLM 67B Chat (V1)OpenDeepSeek	1.0	—	—	—
103	—	DeepSeek R1 Distill Llama 8BOpenDeepSeek	1.0	—	—	—
103	▲ 9	DeepSeek R1 Distill Qwen 1.5BOpenDeepSeek	1.0	—	—	—
103	▲ 4	DeepSeek-Coder-V2OpenDeepSeek	1.0	—	—	—
103	▲ 10	DeepSeek-V2-ChatOpenDeepSeek	1.0	—	—	—
103	▲ 8	Ling-mini-2.0OpenInclusionAI	1.0	—	—	—
103	▲ 15	Qwen Chat 14BOpenAlibaba	1.0	—	—	—
103	▲ 11	Qwen Chat 72BOpenAlibaba	1.0	—	—	—
103	▲ 7	Qwen1.5 Chat 110BOpenAlibaba	1.0	—	—	—
103	▲ 1	Qwen2 Instruct 72BOpenAlibaba	1.0	—	—	—
103	▲ 5	Qwen2.5 Coder Instruct 7BOpenAlibaba	1.0	—	—	—
103	▲ 16	Qwen3 0.6B (Non-reasoning)OpenAlibaba · 1 more variant	1.0	—	—	—
103	▲ 14	Qwen3 1.7B (Non-reasoning)OpenAlibaba · 1 more variant	1.0	—	—	—
103	▲ 3	Qwen3 Omni 30B A3B InstructOpenAlibaba	1.0	$0.43	95	1.8 s
103	▲ 6	Qwen3 VL 4B InstructOpenAlibaba	1.0	—	—	—
103	▲ 2	Qwen3.5 0.8B (Non-reasoning)OpenAlibaba · 1 more variant	1.0	—	—	—

69 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