Model rankings

Rank	Change	Model	Intelligence	Price $/M	Speed tok/s	First token
1	—	GLM-5.2 (max)OpenZ AI	51.1	$2.2	141	1.6 s
2	—	MiniMax-M3OpenMiniMax	44.4	$0.52	88	1.8 s
3	—	DeepSeek V4 Pro (Reasoning, Max Effort)OpenDeepSeek · 2 more variants	44.3	$0.54	82	1.8 s
4	—	Kimi K2.6OpenKimi · 1 more variant	42.8	$1.7	39	2.7 s
5	—	MiMo-V2.5-ProOpenXiaomi · 1 more variant	42.2	$0.54	45	3.2 s
6	—	Kimi K2.7 CodeOpenKimi	41.9	$1.7	63	2.3 s
7	New	Nex-N2-ProOpenNex AGI	41.0	$1.0	117	1.6 s
8	▼ 1	DeepSeek V4 Flash (Reasoning, Max Effort)OpenDeepSeek · 2 more variants	40.3	$0.18	124	1.5 s
9	▼ 1	GLM-5.1 (Reasoning)OpenZ AI · 1 more variant	40.2	$2.2	84	1.5 s
10	▼ 1	MiMo-V2.5OpenXiaomi	40.1	$0.18	88	2.7 s
11	▼ 1	GLM-5 (Reasoning)OpenZ AI · 1 more variant	39.5	$1.6	69	1.8 s
12	▼ 1	MiniMax-M2.7OpenMiniMax	38.1	$0.52	49	2.2 s
13	▼ 1	Kimi K2.5 (Reasoning)OpenKimi · 1 more variant	38.1	$1.2	42	3.0 s
14	▼ 1	Nemotron 3 Ultra 550B A55B (Reasoning)OpenNVIDIA	37.8	$1.2	140	1.5 s
15	▼ 1	Qwen3.6 27B (Reasoning)OpenAlibaba · 1 more variant	37.1	$1.4	56	3.8 s
16	▼ 1	GLM-4.7 (Reasoning)OpenZ AI · 1 more variant	33.8	$1.0	132	0.9 s
17	▼ 1	Qwen3.5 27B (Reasoning)OpenAlibaba · 1 more variant	33.8	$0.83	84	5.8 s
18	▼ 1	Qwen3.5 397B A17B (Reasoning)OpenAlibaba · 1 more variant	33.7	$1.4	52	2.8 s
19	▼ 1	MiniMax-M2.5OpenMiniMax	33.7	$0.52	227	4.5 s
20	▼ 1	Hy3-preview (Reasoning)OpenTencent · 1 more variant	33.6	$0.20	110	3.7 s
21	▼ 1	DeepSeek V3.2 (Reasoning)OpenDeepSeek · 1 more variant	33.4	$0.34	—	—
22	▼ 1	MiMo-V2-Flash (Feb 2026)OpenXiaomi · 2 more variants	33.2	$0.15	93	2.9 s
23	▼ 1	Kimi K2 ThinkingOpenKimi	32.7	$1.1	126	1.4 s
24	▼ 1	Qwen3.5 122B A10B (Reasoning)OpenAlibaba · 1 more variant	32.3	$1.1	141	2.4 s
25	▼ 1	Qwen3.6 35B A3B (Reasoning)OpenAlibaba · 1 more variant	31.6	$0.56	170	2.3 s
26	▼ 1	MiniMax-M2.1OpenMiniMax	31.4	$0.52	215	5.3 s
27	▼ 1	Ring-2.6-1TOpenInclusionAI	30.6	$0.85	128	3.3 s
28	▼ 1	Mistral Medium 3.5OpenMistral	29.9	$3.0	144	1.8 s
29	▼ 1	Step 3.7 FlashOpenStepFun	29.7	$0.44	404	0.9 s
30	▼ 1	Gemma 4 31B (Reasoning)OpenGoogle · 1 more variant	29.4	$0.000	36	1.1 s
31	▼ 1	Command A+OpenCohere · 1 more variant	29.3	$0.000	197	0.4 s
32	▼ 1	Qwen3.5 35B A3B (Reasoning)OpenAlibaba · 1 more variant	29.3	$0.69	174	2.3 s
33	▼ 1	MiniMax-M2OpenMiniMax	28.3	$0.52	116	2.0 s
34	▼ 1	DeepSeek V3.1 Terminus (Reasoning)OpenDeepSeek · 1 more variant	26.3	$1.9	—	—
35	▼ 1	Ling-2.6-1TOpenInclusionAI	26.1	$0.85	—	—
36	▼ 1	Gemma 4 26B A4B (Reasoning)OpenGoogle · 1 more variant	25.7	$0.20	—	—
37	▼ 1	Step 3.5 FlashOpenStepFun	25.5	$0.15	223	1.1 s
38	▼ 1	DeepSeek V3.2 Exp (Reasoning)OpenDeepSeek · 1 more variant	25.4	$0.31	—	—
39	▼ 1	NVIDIA Nemotron 3 Super 120B A12B (Reasoning)OpenNVIDIA	25.4	$0.41	310	1.4 s
40	▼ 1	GLM-4.6 (Reasoning)OpenZ AI · 1 more variant	25.1	$0.96	47	2.5 s
41	▼ 1	Qwen3.5 9B (Reasoning)OpenAlibaba · 1 more variant	25.0	$0.11	57	1.0 s
42	▼ 1	K-EXAONE (Reasoning)OpenLG AI Research · 1 more variant	24.7	—	—	—
43	▼ 1	Trinity Large ThinkingOpenArcee AI	24.5	$0.40	203	1.1 s
44	▼ 1	gpt-oss-120b (high)OpenOpenAI · 1 more variant	23.8	$0.26	301	0.9 s
45	▼ 1	Kimi K2 0905OpenKimi · 1 more variant	23.5	$1.1	27	2.6 s
46	▼ 1	EXAONE 4.5 33BOpenLG AI Research	23.0	—	—	—
47	▼ 1	GLM-4.7-Flash (Reasoning)OpenZ AI · 1 more variant	22.9	$0.15	85	1.3 s
48	▼ 1	Qwen3 235B A22B 2507 (Reasoning)OpenAlibaba · 2 more variants	22.3	$0.84	64	2.8 s
49	▼ 1	DeepSeek V3.2 SpecialeOpenDeepSeek	22.2	—	—	—
50	▼ 1	HyperNova 60B 2605OpenMultiverse Computing	22.1	$0.065	396	0.8 s
51	▼ 1	Gemma 4 12B (Reasoning)OpenGoogle · 1 more variant	22.0	$0.15	126	2.4 s
52	▼ 1	Nemotron Cascade 2 30B A3BOpenNVIDIA	21.3	—	—	—
53	▼ 1	Qwen3 Coder NextOpenAlibaba	21.2	$0.56	108	2.0 s
54	▼ 1	DeepSeek V3.1 (Non-reasoning)OpenDeepSeek · 1 more variant	21.0	$0.83	—	—
55	▼ 1	Mistral Small 4 (Reasoning)OpenMistral · 1 more variant	20.8	$0.26	163	0.8 s
56	▼ 1	Qwen3 VL 235B A22B (Reasoning)OpenAlibaba	20.6	$2.2	55	3.0 s
57	▼ 1	North Mini CodeOpenCohere	20.6	$0.000	117	0.4 s
58	▼ 1	Qwen3.5 4B (Reasoning)OpenAlibaba · 1 more variant	20.1	$0.060	24	1.0 s
59	▼ 1	DeepSeek R1 0528 (May '25)OpenDeepSeek · 1 more variant	20.1	$2.1	—	—
60	▼ 1	Qwen3 Next 80B A3B (Reasoning)OpenAlibaba	19.8	$1.9	188	2.1 s
61	▼ 1	GLM-4.5 (Reasoning)OpenZ AI	19.5	$1.0	—	—
62	▼ 1	Ling 2.6 FlashOpenInclusionAI	19.3	$0.15	196	1.5 s
63	▲ 13	Devstral 2OpenMistral	19.2	$0.000	30	1.7 s
64	▼ 2	Seed-OSS-36B-InstructOpenByteDance Seed	18.3	$0.30	35	3.0 s
65	▼ 2	Qwen3 235B A22B 2507 InstructOpenAlibaba	18.2	$0.36	58	2.5 s
66	▼ 2	Qwen3 Coder 480B A35B InstructOpenAlibaba	18.0	$0.68	59	3.1 s
67	▼ 2	Qwen3 VL 32B (Reasoning)OpenAlibaba	17.9	$2.6	92	2.8 s
68	▼ 2	MiniMax M1 80kOpenMiniMax	17.7	$0.96	—	—
69	▼ 1	LongCat Flash LiteOpenLongCat	17.2	$0.000	—	—
70	▼ 1	GLM-4.6V (Reasoning)OpenZ AI · 1 more variant	16.8	$0.45	76	3.9 s
71	▼ 1	GLM-4.5-AirOpenZ AI	16.5	$0.37	85	2.7 s
72	—	Ring-1TOpenInclusionAI	16.2	—	—	—
73	▼ 2	Mistral Large 3OpenMistral	15.9	$0.75	47	1.2 s
74	▼ 1	Qwen3 30B A3B 2507 (Reasoning)OpenAlibaba · 2 more variants	15.8	$0.67	146	2.6 s
75	▼ 1	DeepSeek V3 0324OpenDeepSeek · 1 more variant	15.7	$1.2	—	—
76	▼ 1	INTELLECT-3OpenPrime Intellect	15.6	—	—	—
77	—	Solar Open 100B (Reasoning)OpenUpstage	15.1	—	—	—
78	—	Nemotron 3 Nano Omni 30B A3B ReasoningOpenNVIDIA	14.9	$0.13	292	1.0 s
79	—	gpt-oss-20B (high)OpenOpenAI · 1 more variant	14.9	$0.088	233	1.0 s
80	—	MiniMax M1 40kOpenMiniMax	14.4	—	—	—
81	—	Qwen3 VL 235B A22B InstructOpenAlibaba	14.3	$0.70	51	2.7 s
82	—	Llama 4 MaverickOpenMeta	14.3	$0.48	104	1.0 s
83	▼ 16	NVIDIA Nemotron 3 Nano 30B A3B (Reasoning)OpenNVIDIA · 1 more variant	14.2	$0.096	75	2.7 s
84	▼ 1	Qwen3 Next 80B A3B InstructOpenAlibaba	13.7	$0.88	187	2.3 s
85	▼ 1	Qwen3 Coder 30B A3B InstructOpenAlibaba	13.6	$0.35	103	2.7 s
86	▼ 1	QwQ 32BOpenAlibaba	13.4	$0.75	32	2.1 s
87	▼ 1	Qwen3 VL 30B A3B (Reasoning)OpenAlibaba	13.3	$0.34	113	2.2 s
88	▼ 1	Devstral Small 2OpenMistral	13.1	$0.000	37	1.6 s
89	▼ 1	Ling-1TOpenInclusionAI	12.8	—	—	—
90	▼ 1	Gemma 4 E4B (Reasoning)OpenGoogle · 1 more variant	12.5	—	—	—
91	▼ 1	Llama Nemotron Super 49B v1.5 (Reasoning)OpenNVIDIA · 1 more variant	12.4	$0.18	50	1.3 s
92	▼ 1	Llama 3.3 Nemotron Super 49B v1 (Reasoning)OpenNVIDIA · 1 more variant	12.2	—	—	—
93	▼ 1	Qwen3 4B 2507 (Reasoning)OpenAlibaba · 2 more variants	12.0	—	—	—
94	▼ 1	MiniCPM5-1B (Reasoning)OpenOpenBMB · 1 more variant	12.0	—	—	—
95	▼ 1	Magistral Small 1.2OpenMistral	11.9	$0.75	81	1.0 s
96	▼ 1	Sarvam 105B (high)OpenSarvam	11.9	$0.074	113	2.1 s
97	▼ 1	Devstral Small (May '25)OpenMistral · 1 more variant	11.8	—	—	—
98	▼ 1	Qwen3 VL 32B InstructOpenAlibaba	11.1	$1.2	68	2.7 s
99	▲ 10	Ministral 3 14BOpenMistral	11.1	$0.20	72	0.9 s
100	▼ 2	DeepSeek R1 Distill Qwen 32BOpenDeepSeek	11.0	—	—	—
101	▼ 2	Magistral Small 1OpenMistral	10.7	—	—	—
102	▼ 2	EXAONE 4.0 32B (Reasoning)OpenLG AI Research · 1 more variant	10.6	—	—	—
103	▼ 2	Qwen3 VL 8B (Reasoning)OpenAlibaba	10.6	$0.66	116	2.4 s
104	▼ 2	Qwen3 32B (Reasoning)OpenAlibaba · 1 more variant	10.5	$0.28	98	2.6 s
105	▼ 2	DeepSeek R1 0528 Qwen3 8BOpenDeepSeek	10.4	—	—	—
106	▼ 2	Qwen3.5 2B (Reasoning)OpenAlibaba · 1 more variant	10.2	$0.040	28	1.0 s
107	▼ 2	Qwen3 14B (Reasoning)OpenAlibaba · 1 more variant	10.1	$0.73	60	2.8 s
108	▼ 2	Nanbeige4.1-3BOpenNanbeige	10.1	—	—	—
109	▼ 2	Llama 4 ScoutOpenMeta	10.0	$0.29	108	0.9 s
110	▼ 2	Qwen3 VL 30B A3B InstructOpenAlibaba	10.0	$0.30	113	2.3 s
111	▼ 1	DeepSeek R1 Distill Llama 70BOpenDeepSeek	9.9	$0.79	68	0.6 s
112	▼ 1	DeepSeek R1 Distill Qwen 14BOpenDeepSeek	9.8	—	—	—
113	▼ 1	Falcon-H1R-7BOpenTII UAE	9.8	—	—	—
114	▼ 1	Ling-flash-2.0OpenInclusionAI	9.7	$0.25	65	2.3 s
115	▼ 1	Qwen3 Omni 30B A3B (Reasoning)OpenAlibaba	9.6	$0.43	88	2.1 s
116	▼ 1	Qwen2.5 Instruct 72BOpenAlibaba	9.6	$0.37	—	—
117	▼ 1	Step3 VL 10BOpenStepFun	9.5	—	—	—
118	▼ 1	Gemma 4 E2B (Reasoning)OpenGoogle · 1 more variant	9.3	—	—	—
119	▼ 1	QwQ 32B-PreviewOpenAlibaba	9.2	—	—	—
120	▼ 1	GLM-4.5V (Reasoning)OpenZ AI · 1 more variant	9.1	$0.90	79	3.8 s
121	▼ 1	Mistral Large 2 (Nov '24)OpenMistral · 1 more variant	9.1	$3.0	48	1.3 s
122	▼ 1	Mistral Small 3.2OpenMistral	9.1	$0.13	144	0.7 s
123	▼ 1	Llama 3.1 Nemotron Ultra 253B v1 (Reasoning)OpenNVIDIA	9.1	$0.90	52	2.4 s
124	▼ 1	Qwen3 30B A3B 2507 InstructOpenAlibaba	9.1	$0.21	169	2.0 s
125	▼ 1	ERNIE 4.5 300B A47BOpenBaidu	9.0	$0.49	—	—
126	▼ 1	NVIDIA Nemotron Nano 12B v2 VL (Reasoning)OpenNVIDIA · 1 more variant	9.0	$0.30	269	0.4 s
127	▼ 1	Ministral 3 8BOpenMistral	9.0	$0.15	82	0.8 s
128	▼ 1	Granite 4.1 30BOpenIBM	8.9	—	—	—
129	▼ 1	NVIDIA Nemotron Nano 9B V2 (Reasoning)OpenNVIDIA · 1 more variant	8.8	$0.070	67	5.2 s
130	▼ 1	NVIDIA Nemotron 3 Nano 4BOpenNVIDIA	8.8	—	—	—
131	▼ 1	Llama 3.3 Instruct 70BOpenMeta	8.6	$0.61	87	1.6 s
132	▼ 1	Mistral Small 3.1OpenMistral	8.6	$0.14	155	0.8 s
133	▼ 1	Llama 3.1 Nemotron Nano 4B v1.1 (Reasoning)OpenNVIDIA	8.5	—	—	—
134	▼ 1	Kimi Linear 48B A3B InstructOpenKimi	8.5	—	—	—
135	▼ 1	Llama 3.1 Instruct 405BOpenMeta	8.5	$3.7	74	2.5 s
136	▼ 1	Qwen3 VL 8B InstructOpenAlibaba	8.4	$0.31	122	2.3 s
137	▼ 1	LFM2.5-8B-A1BOpenLiquid AI	8.3	$0.000	237	4.4 s
138	▼ 1	Ring-flash-2.0OpenInclusionAI	8.2	$0.25	—	—
139	▼ 1	Pixtral LargeOpenMistral	8.1	$3.0	47	1.3 s
140	▼ 1	Grok 2 (Dec '24)OpenxAI	8.0	—	—	—
141	▼ 1	Qwen3 VL 4B (Reasoning)OpenAlibaba	7.9	—	—	—
142	▼ 1	Llama 3.1 Nemotron Instruct 70BOpenNVIDIA	7.6	$1.2	272	4.7 s
143	▼ 1	Qwen2.5 Instruct 32BOpenAlibaba	7.5	—	—	—
144	▼ 1	Qwen3 8B (Reasoning)OpenAlibaba · 1 more variant	7.4	$0.37	38	4.0 s
145	▼ 1	Qwen3 4B 2507 InstructOpenAlibaba	7.1	—	—	—
146	▼ 1	Qwen2.5 Coder Instruct 32BOpenAlibaba	7.1	—	—	—
147	▼ 1	Mistral Small 3OpenMistral	6.9	$0.10	149	0.8 s
148	▼ 1	DeepSeek-V2.5 (Dec '24)OpenDeepSeek · 1 more variant	6.8	—	—	—
149	▲ 9	Ministral 3 3BOpenMistral	6.8	$0.10	145	0.7 s
150	▼ 2	Llama 3.1 Instruct 70BOpenMeta	6.8	$0.56	33	1.9 s
151	▼ 2	Granite 4.1 8BOpenIBM	6.7	$0.063	111	0.9 s
152	▼ 2	Sarvam 30B (high)OpenSarvam	6.6	$0.047	167	1.9 s
153	▼ 2	DeepSeek R1 Distill Llama 8BOpenDeepSeek	6.4	—	—	—
154	▼ 2	R1 1776OpenPerplexity	6.3	—	—	—
155	▼ 2	Llama 3.2 Instruct 90B (Vision)OpenMeta	6.2	$1.4	58	1.2 s
155	▼ 2	Solar MiniOpenUpstage	6.2	$0.15	—	—
157	▼ 2	Llama 3.1 Instruct 8BOpenMeta	6.1	$0.10	153	1.0 s
158	▼ 2	Grok-1OpenxAI	6.0	—	—	—
159	▼ 2	Qwen2 Instruct 72BOpenAlibaba	6.0	—	—	—
160	▼ 1	Jamba 1.7 LargeOpenAI21 Labs	5.3	$3.5	58	1.7 s
161	▼ 1	Granite 4.0 H SmallOpenIBM	5.2	$0.11	428	10 s
162	▼ 1	Jamba 1.5 LargeOpenAI21 Labs	5.1	$3.5	—	—
163	▼ 1	Qwen3 Omni 30B A3B InstructOpenAlibaba	5.1	$0.43	96	2.0 s
164	▼ 1	DeepSeek-Coder-V2OpenDeepSeek	5.1	—	—	—
165	▼ 1	Jamba 1.6 LargeOpenAI21 Labs	5.0	$3.5	60	1.6 s
166	▼ 1	Qwen3.5 0.8B (Reasoning)OpenAlibaba · 1 more variant	5.0	$0.020	26	1.1 s
167	▼ 1	LFM2 24B A2BOpenLiquid AI	4.9	$0.053	128	0.6 s
168	▼ 1	Phi-4OpenMicrosoft	4.9	$0.22	35	2.1 s
169	▼ 1	Gemma 3 27B InstructOpenGoogle	4.8	$0.14	—	—
170	▼ 1	Mistral Small (Sep '24)OpenMistral	4.7	$0.30	159	0.8 s
171	▼ 1	Phi-3 Mini Instruct 3.8BOpenMicrosoft	4.6	—	—	—
172	▼ 1	Gemma 3n E4B Instruct Preview (May '25)OpenGoogle	4.6	—	—	—
173	▼ 1	Phi-4 Multimodal InstructOpenMicrosoft	4.5	$0.000	16	0.9 s
174	▼ 1	Qwen2.5 Coder Instruct 7BOpenAlibaba	4.5	—	—	—
175	▼ 1	Mixtral 8x22B InstructOpenMistral	4.4	—	—	—
176	▼ 1	Llama 2 Chat 7BOpenMeta	4.3	$0.10	—	—
177	▼ 1	Llama 3.2 Instruct 3BOpenMeta	4.2	$0.15	52	1.1 s
178	▼ 1	MiniCPM-V 4.6 1.3BOpenOpenBMB	4.2	—	—	—
179	▼ 1	Jamba Reasoning 3BOpenAI21 Labs	4.1	—	—	—
180	▼ 1	Qwen3 VL 4B InstructOpenAlibaba	4.1	—	—	—
181	▼ 1	Qwen1.5 Chat 110BOpenAlibaba	4.1	—	—	—
182	▼ 1	Reka Flash 3OpenReka AI	4.1	$0.35	—	—
183	▼ 1	Ling-mini-2.0OpenInclusionAI	3.8	—	—	—
184	▼ 1	DeepSeek R1 Distill Qwen 1.5BOpenDeepSeek	3.7	—	—	—
185	▼ 1	DeepSeek-V2-ChatOpenDeepSeek	3.6	—	—	—
186	▼ 1	Llama 3 Instruct 70BOpenMeta	3.5	$1.2	—	—
187	▼ 1	Arctic InstructOpenSnowflake	3.4	—	—	—
187	▼ 1	Qwen Chat 72BOpenAlibaba	3.4	—	—	—
189	▼ 1	Gemma 3 12B InstructOpenGoogle	3.4	$0.14	—	—
190	▼ 1	Llama 3.2 Instruct 11B (Vision)OpenMeta	3.3	$0.25	48	0.8 s
191	▼ 1	Granite 4.1 3BOpenIBM	3.2	—	—	—
192	▼ 1	DeepSeek Coder V2 Lite InstructOpenDeepSeek	3.1	—	—	—
193	▼ 1	Sarvam M (Reasoning)OpenSarvam	3.0	$0.000	—	—
194	▼ 1	Phi-4 Mini InstructOpenMicrosoft	3.0	$0.000	44	1.8 s
195	▼ 1	DeepSeek LLM 67B Chat (V1)OpenDeepSeek	3.0	—	—	—
195	▼ 1	Llama 2 Chat 70BOpenMeta	3.0	—	—	—
197	▼ 1	Llama 2 Chat 13BOpenMeta	3.0	—	—	—
198	▼ 1	Command-R+ (Apr '24)OpenCohere · 1 more variant	3.0	$6.0	—	—
199	▼ 1	OpenChat 3.5 (1210)OpenOpenChat	3.0	—	—	—
200	▼ 1	DBRX InstructOpenDatabricks	3.0	—	—	—
201	▼ 1	Exaone 4.0 1.2B (Reasoning)OpenLG AI Research · 1 more variant	2.9	—	—	—
202	▼ 1	LFM2.5-1.2B-ThinkingOpenLiquid AI	2.7	—	—	—
203	▼ 1	Jamba 1.7 MiniOpenAI21 Labs	2.7	—	—	—
204	▼ 1	LFM2 2.6BOpenLiquid AI	2.7	$0.000	334	1.3 s
205	▼ 1	LFM2.5-1.2B-InstructOpenLiquid AI	2.7	$0.000	465	1.1 s
206	▼ 1	Jamba 1.5 MiniOpenAI21 Labs	2.7	$0.25	—	—
207	▼ 1	Granite 4.0 H 1BOpenIBM	2.7	—	—	—
208	▼ 1	Qwen3 1.7B (Reasoning)OpenAlibaba · 1 more variant	2.6	$0.40	—	—
209	▼ 1	Jamba 1.6 MiniOpenAI21 Labs	2.6	$0.25	183	1.0 s
210	▼ 1	Mixtral 8x7B InstructOpenMistral	2.4	$0.51	—	—
211	▼ 1	Gemma 3 270MOpenGoogle	2.4	—	—	—
212	▼ 1	Apertus 70B InstructOpenSwiss AI Initiative	2.4	$1.3	—	—
213	▼ 1	Granite 4.0 MicroOpenIBM	2.4	—	—	—
214	▼ 1	Llama 65BOpenMeta	2.1	—	—	—
214	▼ 1	Mistral 7B InstructOpenMistral	2.1	$0.21	88	0.9 s
214	▼ 1	Qwen Chat 14BOpenAlibaba	2.1	—	—	—
217	▼ 1	Granite 4.0 1BOpenIBM	2.1	—	—	—
218	▼ 1	LFM2 8B A1BOpenLiquid AI	1.8	$0.000	—	—
219	▼ 1	Granite 3.3 8B (Non-reasoning)OpenIBM	1.8	$0.085	366	27 s
220	▼ 1	Qwen3 0.6B (Reasoning)OpenAlibaba · 1 more variant	1.3	$0.40	—	—
221	▼ 1	Llama 3 Instruct 8BOpenMeta	1.2	$0.070	—	—
222	▼ 1	Gemma 3n E4B InstructOpenGoogle	1.2	$0.025	49	1.2 s
223	▼ 1	LFM2 1.2BOpenLiquid AI	1.1	$0.000	494	1.2 s
224	▼ 1	Gemma 3 4B InstructOpenGoogle	1.1	$0.050	—	—
225	▼ 1	Llama 3.2 Instruct 1BOpenMeta	1.1	$0.050	85	0.9 s
226	▼ 1	LFM2.5-VL-1.6BOpenLiquid AI	1.0	$0.000	467	2.3 s
227	▼ 1	Apertus 8B InstructOpenSwiss AI Initiative	1.0	$0.13	—	—
227	▼ 1	Gemma 3 1B InstructOpenGoogle	1.0	$0.000	—	—
227	▼ 1	Gemma 3n E2B InstructOpenGoogle	1.0	$0.000	—	—
227	▼ 1	Granite 4.0 350MOpenIBM	1.0	—	—	—
227	▼ 1	Granite 4.0 H 350MOpenIBM	1.0	—	—	—
227	▼ 1	Tiny Aya GlobalOpenCohere	1.0	$0.000	—	—

182 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