Model rankings

Rank	Change	Model	Intelligence	Price $/M	Speed tok/s	First token
1	—	Kimi K3 (max)OpenKimi	59.7	$6.0	44	2.4 s
2	—	GLM-5.2 (max)OpenZ AI	52.6	$2.1	130	1.4 s
3	—	DeepSeek V4 Flash 0731 (Reasoning, Max Effort)OpenDeepSeek	51.8	$0.18	136	1.1 s
4	—	Kimi K3 (low)OpenKimi	48.3	$6.0	43	2.8 s
5	—	MiniMax-M3OpenMiniMax	45.4	$0.52	101	1.4 s
6	—	DeepSeek V4 Pro (Reasoning, Max Effort)OpenDeepSeek	45.3	$0.54	73	1.8 s
7	—	Kimi K2.6OpenKimi	45.1	$1.7	54	2.8 s
8	—	DeepSeek V4 Pro (Reasoning, High Effort)OpenDeepSeek	43.7	$0.54	73	1.6 s
9	▲ 1	Kimi K2.7 CodeOpenKimi	43.0	$1.7	42	3.1 s
10	▼ 1	MiMo-V2.5-ProOpenXiaomi	42.9	$0.54	79	2.7 s
11	▲ 2	Inkling (xhigh)OpenThinking Machines	42.3	$1.8	77	1.8 s
12	▼ 1	Hy3OpenTencent	42.2	$0.24	67	2.8 s
13	▲ 1	DeepSeek V4 Flash (Reasoning, Max Effort)OpenDeepSeek	42.1	$0.17	—	—
14	▼ 2	Nex-N2-ProOpenNex AGI	41.7	$1.0	130	1.6 s
15	—	Inkling SmallOpenThinking Machines	41.2	$0.52	144	1.6 s
16	—	GLM-5.1 (Reasoning)OpenZ AI	41.0	$2.1	73	1.4 s
17	—	GLM-5 (Reasoning)OpenZ AI	40.6	$1.6	40	1.2 s
18	▲ 2	DeepSeek V4 Flash (Reasoning, High Effort)OpenDeepSeek	39.0	$0.18	—	—
19	▼ 1	MiniMax-M2.7OpenMiniMax	38.9	$0.52	57	1.8 s
20	▼ 1	Nemotron 3 Ultra 550B A55B (Reasoning)OpenNVIDIA	38.3	$1.1	171	2.1 s
21	—	MiMo-V2.5OpenXiaomi	38.0	$0.18	90	3.0 s
22	New	Ling 3.0 FlashOpenInclusionAI	37.8	$0.11	403	2.1 s
23	▼ 1	Qwen3.6 27B (Reasoning)OpenAlibaba	37.7	$1.4	58	3.8 s
24	—	GLM-5.1 (Non-reasoning)OpenZ AI	36.3	$2.1	63	1.6 s
25	▼ 2	Kimi K2.5 (Reasoning)OpenKimi	36.0	$1.2	71	2.8 s
26	▼ 1	Kimi K2.6 (Non-reasoning)OpenKimi	35.4	$1.7	46	3.0 s
27	▼ 1	GLM-5.2 (Non-reasoning)OpenZ AI	34.8	$2.1	114	1.8 s
28	▼ 1	Qwen3.5 27B (Reasoning)OpenAlibaba	34.6	$0.83	79	5.7 s
29	▲ 1	MiniMax-M2.5OpenMiniMax	34.5	$0.52	105	1.8 s
30	▼ 2	GLM-4.7 (Reasoning)OpenZ AI	34.5	$1.0	112	1.0 s
31	—	Hy3-preview (Reasoning)OpenTencent	34.4	$0.100	—	—
32	▼ 3	Qwen3.5 397B A17B (Reasoning)OpenAlibaba	34.3	$1.4	76	2.4 s
33	—	MiMo-V2-Flash (Feb 2026)OpenXiaomi	34.0	—	—	—
34	▼ 2	LongCat 2.0OpenLongCat	33.9	$1.3	43	2.8 s
35	▼ 1	Kimi K2 ThinkingOpenKimi	33.5	$1.1	129	1.4 s
36	▼ 1	GLM-5 (Non-reasoning)OpenZ AI	33.2	$1.6	39	1.9 s
37	▼ 1	Qwen3.5 122B A10B (Reasoning)OpenAlibaba	32.8	$1.1	135	2.4 s
38	—	Qwen3.5 397B A17B (Non-reasoning)OpenAlibaba	32.7	$1.4	75	2.2 s
39	▼ 2	DeepSeek V3.2 (Reasoning)OpenDeepSeek	32.6	$0.32	—	—
40	▼ 1	Qwen3.6 35B A3B (Reasoning)OpenAlibaba	32.1	$0.56	151	2.3 s
41	▼ 1	MiniMax-M2.1OpenMiniMax	32.1	$0.52	108	1.7 s
42	▼ 1	DeepSeek V4 Pro (Non-reasoning)OpenDeepSeek	31.9	$0.54	72	1.7 s
43	▼ 1	MiMo-V2-Flash (Reasoning)OpenXiaomi	31.9	$0.15	—	—
44	New	G9v3-39A5BOpenAI9Stars	31.4	$0.000	—	—
45	▼ 1	Qwen3.6 27B (Non-reasoning)OpenAlibaba	31.3	$1.4	61	3.9 s
46	▼ 3	Ring-2.6-1TOpenInclusionAI	31.1	$0.85	128	3.3 s
47	▼ 1	Step 3.7 FlashOpenStepFun	30.9	$0.44	397	0.9 s
48	▼ 3	DeepSeek V3.1 Terminus (Reasoning)OpenDeepSeek	30.6	$1.9	—	—
49	▼ 2	Mistral Medium 3.5OpenMistral	30.4	$3.0	136	2.3 s
50	▼ 2	Kimi K2.5 (Non-reasoning)OpenKimi	30.1	$1.2	72	2.8 s
51	▼ 1	Qwen3.5 27B (Non-reasoning)OpenAlibaba	30.0	$0.83	87	5.7 s
52	▼ 1	Qwen3.5 35B A3B (Reasoning)OpenAlibaba	29.9	$0.69	144	2.1 s
53	▼ 4	Gemma 4 31B (Reasoning)OpenGoogle	29.7	$0.000	35	1.1 s
54	▼ 2	GLM-4.6 (Reasoning)OpenZ AI	29.3	$0.96	46	2.4 s
55	▼ 2	DeepSeek V4 Flash (Non-reasoning)OpenDeepSeek	29.3	$0.18	—	—
56	▼ 2	MiniMax-M2OpenMiniMax	28.9	$0.52	107	1.7 s
57	▼ 2	MiMo-V2.5-Pro (Non-reasoning)OpenXiaomi	28.4	$0.54	74	3.1 s
58	▼ 2	Qwen3.5 122B A10B (Non-reasoning)OpenAlibaba	28.2	$1.1	149	2.3 s
59	▼ 2	GLM-4.7 (Non-reasoning)OpenZ AI	27.1	$1.0	112	1.0 s
60	▼ 2	Hy3-preview (Non-reasoning)OpenTencent	26.6	$0.100	—	—
61	▼ 2	Ling-2.6-1TOpenInclusionAI	26.6	$0.85	—	—
62	▼ 2	Gemma 4 26B A4B (Reasoning)OpenGoogle	26.1	$0.20	—	—
63	▼ 2	Step 3.5 FlashOpenStepFun	26.0	$0.15	190	1.1 s
64	▼ 2	DeepSeek V3.2 Exp (Reasoning)OpenDeepSeek	25.9	$0.32	—	—
65	New	Nemotron 3 Super 120B A12B (Reasoning)OpenNVIDIA	25.7	$0.35	138	1.8 s
66	▼ 1	DeepSeek V3.2 (Non-reasoning)OpenDeepSeek	25.1	$0.32	—	—
67	▼ 3	MiMo-V2-Flash (Non-reasoning)OpenXiaomi	25.0	—	—	—
68	▼ 2	Qwen3.6 35B A3B (Non-reasoning)OpenAlibaba	24.5	$0.84	174	2.5 s
69	▼ 2	Qwen3.5 35B A3B (Non-reasoning)OpenAlibaba	24.3	$0.69	170	2.1 s
70	▼ 2	gpt-oss-120b (high)OpenOpenAI	24.1	$0.26	175	0.9 s
71	▼ 2	Kimi K2 0905OpenKimi	24.0	$1.1	37	1.5 s
72	▼ 2	GLM-4.6 (Non-reasoning)OpenZ AI	23.4	$0.98	73	2.8 s
73	▼ 2	GLM-4.7-Flash (Reasoning)OpenZ AI	23.3	$0.15	100	1.6 s
74	▼ 2	Command A+OpenCohere	22.8	$0.000	196	0.4 s
75	▼ 2	DeepSeek V3.2 SpecialeOpenDeepSeek	22.6	—	—	—
76	▼ 2	K-EXAONE (Reasoning)OpenLG AI Research	22.4	—	—	—
77	▼ 2	Gemma 4 12B (Reasoning)OpenGoogle	22.3	$0.15	105	2.4 s
78	▼ 2	Gemma 4 31B (Non-reasoning)OpenGoogle	22.3	$0.21	62	2.2 s
79	▼ 2	Qwen3.5 9B (Reasoning)OpenAlibaba	21.8	$0.15	91	1.6 s
80	▼ 2	DeepSeek V3.1 Terminus (Non-reasoning)OpenDeepSeek	21.7	$0.45	—	—
81	▼ 2	DeepSeek V3.2 Exp (Non-reasoning)OpenDeepSeek	21.7	$0.32	—	—
82	▼ 2	Apriel-v1.5-15B-ThinkerOpenServiceNow	21.6	$0.000	—	—
83	▼ 1	DeepSeek V3.1 (Non-reasoning)OpenDeepSeek	21.4	$0.84	—	—
84	▼ 3	Qwen3 Coder NextOpenAlibaba	21.3	$0.56	148	1.3 s
85	▼ 2	DeepSeek V3.1 (Reasoning)OpenDeepSeek	21.0	$0.87	—	—
86	▼ 2	Qwen3 VL 235B A22B (Reasoning)OpenAlibaba	20.9	$2.6	56	2.8 s
87	▼ 2	Apriel-v1.6-15B-ThinkerOpenServiceNow	20.8	$0.000	—	—
88	▼ 2	Qwen3.5 9B (Non-reasoning)OpenAlibaba	20.6	$0.19	97	0.8 s
89	▼ 2	EXAONE 4.5 33BOpenLG AI Research	20.5	—	—	—
90	▼ 2	Gemma 4 26B A4B (Non-reasoning)OpenGoogle	20.4	$0.20	65	1.0 s
91	▼ 2	Qwen3.5 4B (Reasoning)OpenAlibaba	20.4	$0.060	35	1.0 s
92	▼ 2	DeepSeek R1 0528 (May '25)OpenDeepSeek	20.4	$2.1	—	—
93	▼ 2	North Mini CodeOpenCohere	20.2	$0.000	27	0.6 s
94	▼ 2	Qwen3 235B A22B 2507 (Reasoning)OpenAlibaba	19.9	$2.6	65	2.8 s
95	▼ 1	GLM-4.5 (Reasoning)OpenZ AI	19.7	—	—	—
96	▼ 3	Mistral Small 4 (Reasoning)OpenMistral	19.7	$0.26	163	0.8 s
97	▼ 2	Kimi K2OpenKimi	19.7	$1.0	38	1.5 s
98	▼ 2	Devstral 2OpenMistral	19.2	$0.000	51	1.5 s
99	—	Trinity Large ThinkingOpenArcee AI	18.6	$0.40	198	1.1 s
100	▼ 3	DeepSeek R1 (Jan '25)OpenDeepSeek	18.6	$2.4	—	—
101	▼ 3	Seed-OSS-36B-InstructOpenByteDance Seed	18.5	$0.30	35	3.0 s
102	▼ 2	Qwen3 235B A22B 2507 InstructOpenAlibaba	18.4	$1.2	60	2.3 s
103	—	HyperNova 60B 2605OpenMultiverse Computing	18.3	$0.065	361	0.9 s
104	▼ 3	Qwen3 Coder 480B A35B InstructOpenAlibaba	18.2	$3.0	63	3.0 s
105	▼ 3	Qwen3 VL 32B (Reasoning)OpenAlibaba	18.1	$2.6	88	2.7 s
106	▼ 1	Nemotron Cascade 2 30B A3BOpenNVIDIA	18.0	—	—	—
107	▼ 3	MiniMax M1 80kOpenMiniMax	17.9	$0.96	—	—
108	▼ 2	Devstral Small 2OpenMistral	17.7	$0.000	130	2.3 s
109	▼ 1	LongCat Flash LiteOpenLongCat	17.4	—	—	—
110	▼ 3	K2 Think V2OpenMBZUAI Institute of Foundation Models	17.2	—	—	—
111	▼ 2	HyperCLOVA X SEED Think (32B)OpenNaver	17.2	—	—	—
112	▼ 2	GLM-4.6V (Reasoning)OpenZ AI	16.9	$0.45	58	5.1 s
113	▼ 2	K-EXAONE (Non-reasoning)OpenLG AI Research	16.9	—	—	—
114	▼ 2	Qwen3 Next 80B A3B (Reasoning)OpenAlibaba	16.9	$1.9	205	2.3 s
115	▼ 2	GLM-4.5-AirOpenZ AI	16.7	$0.37	81	2.7 s
116	▼ 1	G9v3-3BOpenAI9Stars	16.5	$0.000	—	—
117	▼ 3	Ring-1TOpenInclusionAI	16.3	—	—	—
118	▼ 2	Qwen3.5 4B (Non-reasoning)OpenAlibaba	16.1	$0.060	37	0.7 s
119	▼ 2	Mistral Large 3OpenMistral	15.9	$0.75	44	1.2 s
120	▼ 2	INTELLECT-3OpenPrime Intellect	15.7	—	—	—
121	▼ 2	GLM-4.7-Flash (Non-reasoning)OpenZ AI	15.6	$0.15	118	1.6 s
122	▼ 1	Solar Open 100B (Reasoning)OpenUpstage	15.2	—	—	—
123	▼ 3	DeepSeek V3 0324OpenDeepSeek	15.2	$0.48	—	—
124	—	gpt-oss-20b (high)OpenOpenAI	15.2	$0.093	134	1.0 s
125	▼ 3	Nemotron 3 Nano Omni 30B A3B ReasoningOpenNVIDIA	15.0	$0.13	321	1.0 s
126	▼ 3	gpt-oss-120b (low)OpenOpenAI	14.9	$0.25	240	0.9 s
127	▼ 2	Mistral Small 3.1OpenMistral	14.9	$0.15	147	0.9 s
128	▼ 2	Qwen3 30B A3B 2507 (Reasoning)OpenAlibaba	14.6	$0.75	145	2.5 s
129	▲ 3	NVIDIA Nemotron 3 Nano 30B A3B (Reasoning)OpenNVIDIA	14.5	$0.088	239	1.2 s
130	—	Llama 4 MaverickOpenMeta	14.5	$0.42	136	1.0 s
131	▼ 4	MiniMax M1 40kOpenMiniMax	14.5	—	—	—
132	▼ 4	gpt-oss-20b (low)OpenOpenAI	14.4	$0.10	141	1.1 s
133	▼ 4	Qwen3 VL 235B A22B InstructOpenAlibaba	14.4	$1.2	47	2.7 s
134	▼ 3	K2-V2 (high)OpenMBZUAI Institute of Foundation Models	14.2	—	—	—
135	▼ 2	DeepSeek V3 (Dec '24)OpenDeepSeek	14.2	$0.49	—	—
136	▼ 2	Ling 2.6 FlashOpenInclusionAI	14.2	$0.15	110	1.1 s
137	▲ 1	DiffusionGemma 26B A4BOpenGoogle	13.8	—	—	—
138	▼ 3	Qwen3 Next 80B A3B InstructOpenAlibaba	13.8	$0.88	200	2.2 s
139	▼ 3	Tri-21B-think PreviewOpenTrillion Labs	13.6	—	—	—
140	▼ 3	Qwen3 Coder 30B A3B InstructOpenAlibaba	13.6	$0.90	94	2.6 s
141	▼ 2	Qwen3 235B A22B (Reasoning)OpenAlibaba	13.5	$2.6	57	2.9 s
142	▼ 2	QwQ 32BOpenAlibaba	13.4	$0.75	32	2.1 s
143	▼ 2	Qwen3 VL 30B A3B (Reasoning)OpenAlibaba	13.4	$0.75	106	2.2 s
144	▼ 2	Gemma 4 12B (Non-reasoning)OpenGoogle	13.2	$0.15	107	2.5 s
145	▼ 2	Ling-1TOpenInclusionAI	12.7	—	—	—
146	▼ 2	K2-V2 (medium)OpenMBZUAI Institute of Foundation Models	12.4	—	—	—
147	▼ 2	Llama Nemotron Super 49B v1.5 (Reasoning)OpenNVIDIA	12.4	$0.40	130	4.2 s
148	▲ 1	Gemma 4 E4B (Reasoning)OpenGoogle	12.4	$0.040	77	0.9 s
149	▼ 3	Mistral Small 4 (Non-reasoning)OpenMistral	12.3	$0.26	145	0.8 s
150	▼ 3	Tri-21B-ThinkOpenTrillion Labs	12.3	—	—	—
151	▼ 3	Llama 3.3 Nemotron Super 49B v1 (Reasoning)OpenNVIDIA	12.2	—	—	—
152	▼ 2	Qwen3 4B 2507 (Reasoning)OpenAlibaba	11.9	—	—	—
153	▼ 2	MiniCPM5-1B (Reasoning)OpenOpenBMB	11.9	—	—	—
154	▼ 2	Sarvam 105B (high)OpenSarvam	11.9	$0.074	—	—
155	▼ 2	Devstral Small (May '25)OpenMistral	11.8	—	—	—
156	▼ 2	MiniCPM5-1B (Non-reasoning)OpenOpenBMB	11.7	—	—	—
157	▼ 1	Magistral Small 1.2OpenMistral	11.5	$0.75	144	0.8 s
158	▼ 3	Qwen3 32B (Reasoning)OpenAlibaba	11.4	$2.6	91	2.5 s
159	—	Ministral 3 14BOpenMistral	11.2	$0.20	99	0.9 s
160	▼ 3	Nanbeige4.1-3BOpenNanbeige	11.2	—	—	—
161	▼ 3	Qwen3 VL 32B InstructOpenAlibaba	11.0	$1.2	67	2.8 s
162	▼ 2	DeepSeek R1 Distill Qwen 32BOpenDeepSeek	11.0	—	—	—
163	▼ 2	GLM-4.6V (Non-reasoning)OpenZ AI	10.9	$0.45	71	4.2 s
164	▼ 2	Qwen3 235B A22B (Non-reasoning)OpenAlibaba	10.8	$1.2	60	2.7 s
165	▲ 1	Mistral Small 3.2OpenMistral	10.7	$0.15	152	0.8 s
166	▼ 3	Magistral Small 1OpenMistral	10.6	—	—	—
167	▼ 3	EXAONE 4.0 32B (Reasoning)OpenLG AI Research	10.5	—	—	—
168	▼ 3	Qwen3 VL 8B (Reasoning)OpenAlibaba	10.5	$0.66	109	2.4 s
169	▼ 2	Qwen3 14B (Reasoning)OpenAlibaba	10.4	$1.3	58	3.0 s
170	▼ 2	DeepSeek R1 0528 Qwen3 8BOpenDeepSeek	10.3	—	—	—
171	▼ 2	Llama 4 ScoutOpenMeta	10.3	$0.30	105	0.8 s
172	▼ 2	Qwen3 VL 30B A3B InstructOpenAlibaba	9.9	$0.35	110	2.2 s
173	▼ 2	Hermes 4 - Llama-3.1 70B (Reasoning)OpenNous Research	9.9	$0.20	89	1.4 s
174	▼ 2	DeepSeek R1 Distill Llama 70BOpenDeepSeek	9.8	$0.80	31	0.8 s
175	▲ 1	Gemma 4 E2B (Reasoning)OpenGoogle	9.8	—	—	—
176	▼ 3	DeepSeek R1 Distill Qwen 14BOpenDeepSeek	9.7	—	—	—
177	▼ 3	Falcon-H1R-7BOpenTII UAE	9.7	—	—	—
178	▼ 3	Ling-flash-2.0OpenInclusionAI	9.6	$0.25	89	2.3 s
179	▼ 2	Qwen3 Omni 30B A3B (Reasoning)OpenAlibaba	9.5	$0.43	94	2.0 s
180	▼ 2	Qwen2.5 Instruct 72BOpenAlibaba	9.4	$0.48	—	—
181	▼ 2	Step3 VL 10BOpenStepFun	9.3	—	—	—
182	▼ 2	Llama 3.3 Instruct 70BOpenMeta	9.3	$0.67	81	1.7 s
183	▼ 2	Qwen3 30B A3B (Reasoning)OpenAlibaba	9.2	$0.75	101	2.2 s
184	▼ 2	Devstral Small (Jul '25)OpenMistral	9.1	—	—	—
185	▼ 2	QwQ 32B-PreviewOpenAlibaba	9.1	—	—	—
186	▼ 2	GLM-4.5V (Reasoning)OpenZ AI	9.0	$0.90	98	1.9 s
187	▼ 2	Mistral Large 2 (Nov '24)OpenMistral	9.0	—	—	—
188	▲ 3	Ministral 3 8BOpenMistral	9.0	$0.15	128	0.7 s
189	▲ 4	Granite 4.1 30BOpenIBM	8.9	—	—	—
190	▼ 4	Llama 3.1 Nemotron Ultra 253B v1 (Reasoning)OpenNVIDIA	8.9	$0.90	52	2.4 s
191	▼ 4	Qwen3 30B A3B 2507 InstructOpenAlibaba	8.9	$0.35	147	2.0 s
192	▼ 4	ERNIE 4.5 300B A47BOpenBaidu	8.9	$0.49	—	—
193	▼ 4	Hermes 4 - Llama-3.1 405B (Reasoning)OpenNous Research	8.8	$1.5	42	2.4 s
194	▲ 2	NVIDIA Nemotron 3 Nano 4BOpenNVIDIA	8.8	—	—	—
195	▼ 5	NVIDIA Nemotron Nano 12B v2 VL (Reasoning)OpenNVIDIA	8.8	$0.30	100	4.8 s
196	▼ 4	Gemma 4 E4B (Non-reasoning)OpenGoogle	8.7	$0.040	79	0.8 s
197	▼ 3	NVIDIA Nemotron Nano 9B V2 (Reasoning)OpenNVIDIA	8.7	$0.070	147	4.6 s
198	▼ 3	Hermes 4 - Llama-3.1 405B (Non-reasoning)OpenNous Research	8.6	$1.5	42	2.3 s
199	▼ 2	Llama Nemotron Super 49B v1.5 (Non-reasoning)OpenNVIDIA	8.5	$0.40	106	6.0 s
200	▼ 2	Qwen3 32B (Non-reasoning)OpenAlibaba	8.5	$1.2	93	2.4 s
201	▼ 2	K2-V2 (low)OpenMBZUAI Institute of Foundation Models	8.4	—	—	—
202	▼ 2	Llama 3.1 Nemotron Nano 4B v1.1 (Reasoning)OpenNVIDIA	8.4	—	—	—
203	▼ 2	Kimi Linear 48B A3B InstructOpenKimi	8.4	—	—	—
204	▼ 2	Llama 3.1 Instruct 405BOpenMeta	8.3	—	—	—
205	▲ 2	Qwen3 8B (Reasoning)OpenAlibaba	8.3	$0.66	38	3.8 s
206	▼ 3	Llama 3.3 Nemotron Super 49B v1 (Non-reasoning)OpenNVIDIA	8.3	—	—	—
207	▼ 3	Qwen3 VL 8B InstructOpenAlibaba	8.2	$0.31	110	2.2 s
208	▼ 3	Qwen3 4B (Reasoning)OpenAlibaba	8.2	—	—	—
209	▼ 3	LFM2.5-8B-A1BOpenLiquid AI	8.1	$0.000	333	1.9 s
210	▼ 2	Llama 3.1 Tulu3 405BOpenAllen Institute for AI	8.1	—	—	—
211	▼ 2	Ring-flash-2.0OpenInclusionAI	8.0	$0.25	—	—
212	▼ 2	Pixtral LargeOpenMistral	8.0	—	—	—
213	▼ 2	Olmo 3.1 32B ThinkOpenAllen Institute for AI	7.9	$0.000	—	—
214	▼ 2	Grok 2 (Dec '24)OpenSpaceXAI	7.8	—	—	—
215	▲ 2	Llama 3.1 Instruct 8BOpenMeta	7.8	$0.028	143	0.9 s
216	▼ 3	Qwen3 VL 4B (Reasoning)OpenAlibaba	7.7	—	—	—
217	▼ 2	Qwen3.5 2B (Reasoning)OpenAlibaba	7.5	—	—	—
218	▼ 4	Command AOpenCohere	7.5	$4.4	61	1.8 s
219	▼ 3	Llama 3.1 Nemotron Instruct 70BOpenNVIDIA	7.4	$1.2	114	5.0 s
220	▼ 1	Gemma 3 27B InstructOpenGoogle	7.4	$0.000	—	—
221	▼ 3	Qwen2.5 Instruct 32BOpenAlibaba	7.2	—	—	—
222	▼ 2	NVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning)OpenNVIDIA	7.2	$0.088	152	0.6 s
223	▼ 2	NVIDIA Nemotron Nano 9B V2 (Non-reasoning)OpenNVIDIA	7.2	$0.086	157	1.8 s
224	▲ 6	Ministral 3 3BOpenMistral	7.1	$0.10	210	0.6 s
225	▼ 3	Mistral Large 2 (Jul '24)OpenMistral	7.0	$3.0	—	—
226	▼ 3	Qwen3 4B 2507 InstructOpenAlibaba	6.9	—	—	—
227	▼ 3	Qwen2.5 Coder Instruct 32BOpenAlibaba	6.9	—	—	—
228	▼ 3	Qwen3 14B (Non-reasoning)OpenAlibaba	6.8	$0.61	59	2.8 s
229	▼ 3	GLM-4.5V (Non-reasoning)OpenZ AI	6.8	$0.90	97	1.8 s
230	▼ 3	Mistral Small 3OpenMistral	6.7	$0.15	144	0.9 s
231	▼ 3	Hermes 4 - Llama-3.1 70B (Non-reasoning)OpenNous Research	6.7	$0.20	90	1.4 s
232	▼ 3	Qwen3 30B A3B (Non-reasoning)OpenAlibaba	6.6	$0.35	104	2.2 s
233	▼ 2	DeepSeek-V2.5 (Dec '24)OpenDeepSeek	6.5	—	—	—
234	▼ 2	Qwen3 4B (Non-reasoning)OpenAlibaba	6.5	—	—	—
235	▼ 2	Llama 3.1 Instruct 70BOpenMeta	6.5	$0.56	64	1.6 s
236	▼ 2	Granite 4.1 8BOpenIBM	6.4	$0.063	86	0.8 s
237	▼ 2	Sarvam 30B (high)OpenSarvam	6.4	$0.047	—	—
238	▼ 2	DeepSeek-V2.5OpenDeepSeek	6.4	—	—	—
239	▲ 7	Phi-4 Mini InstructOpenMicrosoft	6.2	$0.000	43	0.8 s
240	▼ 3	Olmo 3.1 32B InstructOpenAllen Institute for AI	6.2	—	—	—
241	▼ 3	DeepSeek R1 Distill Llama 8BOpenDeepSeek	6.2	—	—	—
242	▼ 3	Gemma 4 E2B (Non-reasoning)OpenGoogle	6.2	—	—	—
243	▼ 3	Olmo 3 32B ThinkOpenAllen Institute for AI	6.1	—	—	—
244	▼ 3	R1 1776OpenPerplexity	6.0	—	—	—
245	▼ 3	Llama 3.2 Instruct 90B (Vision)OpenMeta	6.0	—	—	—
245	▼ 3	Solar MiniOpenUpstage	6.0	$0.15	—	—
247	▼ 3	Grok-1OpenSpaceXAI	5.8	—	—	—
248	▼ 3	Qwen2 Instruct 72BOpenAlibaba	5.7	—	—	—
249	▼ 2	EXAONE 4.0 32B (Non-reasoning)OpenLG AI Research	5.7	—	—	—
250	▼ 2	Qwen3.5 2B (Non-reasoning)OpenAlibaba	5.7	—	—	—
251	▼ 2	Gemma 3 12B InstructOpenGoogle	5.5	$0.000	—	—
252	▼ 2	Qwen3.5 0.8B (Reasoning)OpenAlibaba	5.0	—	—	—
253	▼ 2	DeepHermes 3 - Mistral 24B Preview (Non-reasoning)OpenNous Research	5.0	—	—	—
254	▼ 2	Jamba 1.7 LargeOpenAI21 Labs	5.0	$3.5	54	1.3 s
255	▼ 2	Granite 4.0 H SmallOpenIBM	4.9	$0.11	45	11 s
256	▼ 2	Jamba 1.5 LargeOpenAI21 Labs	4.8	$3.5	—	—
257	▼ 2	Qwen3 Omni 30B A3B InstructOpenAlibaba	4.8	$0.43	93	1.9 s
258	▼ 2	Hermes 3 - Llama-3.1 70BOpenNous Research	4.8	$0.70	31	2.0 s
259	▼ 2	Qwen3 8B (Non-reasoning)OpenAlibaba	4.8	$0.31	40	3.7 s
260	▲ 3	Granite 4.1 3BOpenIBM	4.7	—	—	—
261	▼ 3	DeepSeek-Coder-V2OpenDeepSeek	4.7	—	—	—
262	▼ 3	OLMo 2 32BOpenAllen Institute for AI	4.7	—	—	—
263	▼ 3	Jamba 1.6 LargeOpenAI21 Labs	4.7	$3.5	56	1.3 s
264	▼ 3	LFM2 24B A2BOpenLiquid AI	4.6	—	—	—
265	▼ 3	Phi-4OpenMicrosoft	4.6	$0.22	—	—
266	▼ 2	Mistral Small (Sep '24)OpenMistral	4.3	$0.30	152	0.8 s
267	▲ 6	MiniCPM-V 4.6 1.3BOpenOpenBMB	4.3	—	—	—
268	▼ 3	Phi-3 Mini Instruct 3.8BOpenMicrosoft	4.3	—	—	—
269	▼ 3	NVIDIA Nemotron Nano 12B v2 VL (Non-reasoning)OpenNVIDIA	4.2	$0.30	196	1.8 s
270	▼ 3	Gemma 3n E4B Instruct Preview (May '25)OpenGoogle	4.2	—	—	—
271	▼ 3	Phi-4 Multimodal InstructOpenMicrosoft	4.2	$0.000	18	0.8 s
272	▼ 3	Qwen2.5 Coder Instruct 7BOpenAlibaba	4.1	—	—	—
273	▼ 3	Mixtral 8x22B InstructOpenMistral	4.0	—	—	—
274	▼ 3	Llama 2 Chat 7BOpenMeta	3.9	$0.10	—	—
275	▼ 3	Llama 3.2 Instruct 3BOpenMeta	3.9	—	—	—
276	▼ 2	Jamba Reasoning 3BOpenAI21 Labs	3.8	—	—	—
277	▼ 2	Qwen3 VL 4B InstructOpenAlibaba	3.7	—	—	—
278	▼ 2	Qwen1.5 Chat 110BOpenAlibaba	3.7	—	—	—
279	▼ 2	Reka Flash 3OpenReka AI	3.7	$0.35	—	—
280	▼ 2	Olmo 3 7B ThinkOpenAllen Institute for AI	3.6	—	—	—
281	▲ 7	Qwen3.5 0.8B (Non-reasoning)OpenAlibaba	3.5	—	—	—
282	▼ 3	OLMo 2 7BOpenAllen Institute for AI	3.5	—	—	—
283	▼ 3	Molmo 7B-DOpenAllen Institute for AI	3.4	—	—	—
284	▼ 3	Ling-mini-2.0OpenInclusionAI	3.4	—	—	—
285	▼ 3	DeepSeek R1 Distill Qwen 1.5BOpenDeepSeek	3.3	—	—	—
286	▼ 3	DeepSeek-V2-ChatOpenDeepSeek	3.3	—	—	—
287	▼ 3	Llama 3 Instruct 70BOpenMeta	3.1	$1.2	—	—
288	▼ 3	Arctic InstructOpenSnowflake	3.0	—	—	—
288	▼ 3	Qwen Chat 72BOpenAlibaba	3.0	—	—	—
290	▼ 3	Llama 3.2 Instruct 11B (Vision)OpenMeta	3.0	$0.35	26	1.7 s
291	▼ 2	DeepSeek Coder V2 Lite InstructOpenDeepSeek	2.7	—	—	—
292	▼ 2	Sarvam M (Reasoning)OpenSarvam	2.6	$0.000	—	—
293	▼ 2	DeepSeek LLM 67B Chat (V1)OpenDeepSeek	2.6	—	—	—
293	▼ 2	Llama 2 Chat 70BOpenMeta	2.6	—	—	—
295	▼ 2	Llama 2 Chat 13BOpenMeta	2.6	—	—	—
296	▼ 2	Command-R+ (Apr '24)OpenCohere	2.6	$6.0	—	—
297	▼ 2	OpenChat 3.5 (1210)OpenOpenChat	2.6	—	—	—
298	▼ 2	DBRX InstructOpenDatabricks	2.6	—	—	—
299	▼ 2	Exaone 4.0 1.2B (Reasoning)OpenLG AI Research	2.5	—	—	—
300	▼ 2	Olmo 3 7B InstructOpenAllen Institute for AI	2.4	$0.13	—	—
301	▼ 2	Exaone 4.0 1.2B (Non-reasoning)OpenLG AI Research	2.4	—	—	—
302	▼ 2	LFM2.5-1.2B-ThinkingOpenLiquid AI	2.3	—	—	—
303	▼ 2	Jamba 1.7 MiniOpenAI21 Labs	2.3	—	—	—
304	▼ 2	LFM2 2.6BOpenLiquid AI	2.3	—	—	—
305	▼ 2	LFM2.5-1.2B-InstructOpenLiquid AI	2.3	—	—	—
306	▼ 2	Jamba 1.5 MiniOpenAI21 Labs	2.3	$0.25	—	—
307	▼ 2	Granite 4.0 H 1BOpenIBM	2.2	—	—	—
308	▼ 2	Qwen3 1.7B (Reasoning)OpenAlibaba	2.2	—	—	—
309	▼ 2	Jamba 1.6 MiniOpenAI21 Labs	2.1	$0.25	180	0.8 s
310	▼ 2	Mixtral 8x7B InstructOpenMistral	2.0	$0.51	—	—
311	▼ 2	Gemma 3 270MOpenGoogle	2.0	—	—	—
312	▼ 2	Apertus 70B InstructOpenSwiss AI Initiative	2.0	$1.3	—	—
313	▼ 2	Granite 4.0 MicroOpenIBM	2.0	—	—	—
314	▼ 2	DeepHermes 3 - Llama-3.1 8B Preview (Non-reasoning)OpenNous Research	1.9	—	—	—
315	▼ 2	Command-R (Mar '24)OpenCohere	1.7	$0.75	—	—
315	▼ 2	Llama 65BOpenMeta	1.7	—	—	—
315	▼ 2	Mistral 7B InstructOpenMistral	1.7	$0.25	125	0.8 s
315	▼ 2	Qwen Chat 14BOpenAlibaba	1.7	—	—	—
319	▼ 2	Granite 4.0 1BOpenIBM	1.6	—	—	—
320	▼ 2	Molmo2-8BOpenAllen Institute for AI	1.6	—	—	—
321	▼ 2	LFM2 8B A1BOpenLiquid AI	1.3	—	—	—
322	▼ 2	Granite 3.3 8B (Non-reasoning)OpenIBM	1.3	$0.085	15	27 s
323	▼ 2	Qwen3 1.7B (Non-reasoning)OpenAlibaba	1.1	—	—	—
324	▲ 5	Apertus 8B InstructOpenSwiss AI Initiative	1.0	$0.13	—	—
324	▲ 5	Gemma 3 1B InstructOpenGoogle	1.0	$0.000	—	—
324	▲ 2	Gemma 3 4B InstructOpenGoogle	1.0	$0.000	—	—
324	▲ 5	Gemma 3n E2B InstructOpenGoogle	1.0	$0.000	—	—
324	—	Gemma 3n E4B InstructOpenGoogle	1.0	$0.075	49	1.4 s
324	▲ 5	Granite 4.0 350MOpenIBM	1.0	—	—	—
324	▲ 5	Granite 4.0 H 350MOpenIBM	1.0	—	—	—
324	▲ 1	LFM2 1.2BOpenLiquid AI	1.0	—	—	—
324	▲ 4	LFM2.5-VL-1.6BOpenLiquid AI	1.0	$0.000	353	1.0 s
324	▼ 1	Llama 3 Instruct 8BOpenMeta	1.0	$0.070	—	—
324	▲ 3	Llama 3.2 Instruct 1BOpenMeta	1.0	—	—	—
324	▲ 5	Qwen3 0.6B (Non-reasoning)OpenAlibaba	1.0	—	—	—
324	▼ 2	Qwen3 0.6B (Reasoning)OpenAlibaba	1.0	—	—	—
324	▲ 5	Tiny Aya GlobalOpenCohere	1.0	$0.000	—	—

287 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