Model rankings

Rank	Change	Model	Intelligence	Price $/M	Speed tok/s	First token
1	New	Claude Opus 4.8 (Adaptive Reasoning, Max Effort)ClosedAnthropic	61.4	$11	64	20 s
2	▼ 1	GPT-5.5 (xhigh)ClosedOpenAI · 4 more variants	60.2	$11	60	64 s
3	▼ 1	Claude Opus 4.7 (Adaptive Reasoning, Max Effort)ClosedAnthropic · 1 more variant	57.3	$11	55	16 s
4	▼ 1	Gemini 3.1 Pro PreviewClosedGoogle	57.2	$4.5	133	22 s
5	▼ 1	GPT-5.4 (xhigh)ClosedOpenAI · 2 more variants	56.8	$5.6	84	157 s
6	▼ 1	Qwen3.7 MaxClosedAlibaba	56.6	$3.8	201	2.5 s
7	▼ 1	Gemini 3.5 Flash (high)ClosedGoogle · 2 more variants	55.3	$3.4	206	17 s
8	▼ 1	Kimi K2.6OpenKimi · 1 more variant	53.9	$1.7	41	3.0 s
9	▼ 1	MiMo-V2.5-ProOpenXiaomi · 1 more variant	53.8	$0.54	47	3.7 s
10	▼ 1	GPT-5.3 Codex (xhigh)ClosedOpenAI	53.6	$4.8	74	125 s
11	▼ 1	Grok 4.3 (high)ClosedxAI · 3 more variants	53.2	$1.6	176	12 s
12	▼ 1	Claude Opus 4.6 (Adaptive Reasoning, Max Effort)ClosedAnthropic · 1 more variant	52.9	$11	42	15 s
13	▼ 1	Muse SparkClosedMeta	52.2	—	—	—
14	▼ 1	Qwen3.6 Max PreviewClosedAlibaba	51.8	$2.9	35	3.7 s
15	▼ 1	Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort)ClosedAnthropic · 2 more variants	51.7	$6.6	61	111 s
16	▼ 1	DeepSeek V4 Pro (Reasoning, Max Effort)OpenDeepSeek · 2 more variants	51.5	$0.54	49	1.8 s
17	▼ 1	GLM-5.1 (Reasoning)OpenZ AI · 1 more variant	51.4	$2.2	58	1.6 s
18	▼ 1	GPT-5.2 (xhigh)ClosedOpenAI · 2 more variants	51.3	$4.8	72	122 s
19	▼ 1	Qwen3.6 PlusClosedAlibaba	50.0	$1.1	53	3.1 s
20	▼ 1	GLM-5 (Reasoning)OpenZ AI · 1 more variant	49.8	$1.6	70	1.4 s
21	▼ 1	Claude Opus 4.5 (Reasoning)ClosedAnthropic · 1 more variant	49.7	$11	48	17 s
22	▼ 1	MiniMax-M2.7OpenMiniMax	49.6	$0.52	73	2.5 s
23	▼ 1	Grok 4.20 0309 v2 (Reasoning)ClosedxAI · 1 more variant	49.3	$3.0	185	13 s
24	▼ 1	MiMo-V2-ProClosedXiaomi	49.2	$1.5	59	3.0 s
25	▼ 1	GPT-5.2 Codex (xhigh)ClosedOpenAI	49.0	$4.8	105	26 s
26	▼ 1	MiMo-V2.5OpenXiaomi	49.0	$0.18	86	3.7 s
27	▼ 1	GPT-5.4 mini (xhigh)ClosedOpenAI · 2 more variants	48.9	$1.7	152	7.5 s
28	▼ 1	Grok 4.20 0309 (Reasoning)ClosedxAI · 1 more variant	48.5	$3.0	187	11 s
29	▼ 1	Gemini 3 Pro Preview (high)ClosedGoogle · 1 more variant	48.4	$4.5	135	44 s
30	▼ 1	GPT-5.1 (high)ClosedOpenAI · 1 more variant	47.7	$3.4	97	37 s
31	▼ 1	Kimi K2.5 (Reasoning)OpenKimi · 1 more variant	46.8	$1.2	46	3.0 s
32	▼ 1	GLM-5-TurboClosedZ AI	46.8	—	—	—
33	▼ 1	DeepSeek V4 Flash (Reasoning, Max Effort)OpenDeepSeek · 2 more variants	46.5	$0.18	104	1.2 s
34	▼ 1	Gemini 3 Flash Preview (Reasoning)ClosedGoogle · 1 more variant	46.4	$1.1	173	7.5 s
35	▼ 1	Qwen3.6 27B (Reasoning)OpenAlibaba · 1 more variant	45.8	$1.4	59	3.7 s
36	▼ 1	Qwen3.5 397B A17B (Reasoning)OpenAlibaba · 1 more variant	45.0	$1.4	52	2.5 s
37	▼ 1	MiMo-V2-Omni-0327ClosedXiaomi · 1 more variant	44.9	$0.80	105	3.5 s
38	▼ 1	GPT-5 Codex (high)ClosedOpenAI	44.6	$3.4	189	7.9 s
39	▼ 1	GPT-5 (high)ClosedOpenAI · 4 more variants	44.6	$3.4	88	76 s
40	▼ 1	GPT-5.4 nano (xhigh)ClosedOpenAI · 2 more variants	44.0	$0.46	142	4.2 s
41	▼ 1	Qwen3.6 35B A3B (Reasoning)OpenAlibaba · 1 more variant	43.5	$0.56	180	2.5 s
42	▼ 1	GPT-5.1 Codex (high)ClosedOpenAI	43.1	$3.4	178	6.4 s
43	▼ 1	Claude 4.5 Sonnet (Reasoning)ClosedAnthropic · 1 more variant	43.0	$6.6	42	11 s
44	▼ 1	GLM 5V Turbo (Reasoning)ClosedZ AI	42.9	—	—	—
45	▼ 1	GLM-4.7 (Reasoning)OpenZ AI · 1 more variant	42.1	$1.0	101	1.4 s
46	▼ 1	Qwen3.5 27B (Reasoning)OpenAlibaba · 1 more variant	42.1	$0.83	81	5.7 s
47	▼ 1	Claude 4.1 Opus (Reasoning)ClosedAnthropic · 1 more variant	42.0	$33	30	14 s
48	▼ 1	MiniMax-M2.5OpenMiniMax	41.9	$0.52	148	3.0 s
49	▼ 1	Hy3-preview (Reasoning)OpenTencent · 1 more variant	41.9	$0.20	98	4.0 s
50	New	GPT-5.5 Instant (May 2026)ClosedOpenAI	41.8	$11	—	—
51	▼ 2	DeepSeek V3.2 (Reasoning)OpenDeepSeek · 1 more variant	41.7	$0.34	—	—
52	▼ 2	Qwen3.5 122B A10B (Reasoning)OpenAlibaba · 1 more variant	41.6	$1.1	140	2.4 s
53	▼ 2	Grok 4ClosedxAI	41.5	$11	—	—
54	▼ 2	MiMo-V2-Flash (Feb 2026)OpenXiaomi · 2 more variants	41.5	$0.15	130	2.3 s
55	▼ 2	GPT-5 mini (high)ClosedOpenAI · 2 more variants	41.2	$0.69	88	94 s
56	▼ 2	Kimi K2 ThinkingOpenKimi	40.9	$1.1	116	1.5 s
57	▼ 2	o3-proClosedOpenAI	40.7	$35	29	91 s
58	▼ 2	Qwen3 Max ThinkingClosedAlibaba · 1 more variant	39.8	$2.4	44	4.1 s
59	▼ 2	MiniMax-M2.1OpenMiniMax	39.4	$0.52	125	3.0 s
60	▼ 2	Mistral Medium 3.5OpenMistral	39.2	$3.0	152	1.6 s
61	▼ 2	Gemma 4 31B (Reasoning)OpenGoogle · 1 more variant	39.2	$0.000	35	1.1 s
62	▼ 2	Claude 4 Opus (Reasoning)ClosedAnthropic · 1 more variant	39.0	$33	30	8.3 s
63	▼ 2	Claude 4 Sonnet (Reasoning)ClosedAnthropic · 1 more variant	38.7	$6.6	42	13 s
64	▼ 2	Qwen3.5 Omni PlusClosedAlibaba	38.6	$1.5	54	2.5 s
65	▼ 2	GPT-5.1 Codex mini (high)ClosedOpenAI	38.6	$0.69	171	6.2 s
66	▼ 2	Grok 4.1 Fast (Reasoning)ClosedxAI · 1 more variant	38.6	—	—	—
67	▼ 2	Step 3.5 Flash 2603ClosedStepFun · 1 more variant	38.5	$0.000	174	1.1 s
68	▼ 2	Ring-2.6-1TOpenInclusionAI	38.5	$0.85	125	3.3 s
69	▼ 2	o3ClosedOpenAI	38.4	$3.5	110	6.3 s
70	▼ 2	Command A+OpenCohere · 1 more variant	37.2	$0.000	223	0.2 s
71	▼ 2	Qwen3.5 35B A3B (Reasoning)OpenAlibaba · 1 more variant	37.1	$0.69	148	2.4 s
72	▼ 2	Claude 4.5 Haiku (Reasoning)ClosedAnthropic · 1 more variant	37.1	$2.2	97	17 s
73	▼ 2	JT-35B-FlashClosedChina Mobile	36.1	—	—	—
74	▼ 2	MiniMax-M2OpenMiniMax	36.1	$0.52	92	1.8 s
75	▼ 2	NVIDIA Nemotron 3 Super 120B A12B (Reasoning)OpenNVIDIA	36.0	$0.41	153	1.8 s
76	▼ 2	Nova 2.0 Pro Preview (medium)ClosedAmazon · 2 more variants	35.7	$3.4	119	12 s
77	▼ 2	Grok 4 Fast (Reasoning)ClosedxAI · 1 more variant	35.1	$0.28	—	—
78	▼ 2	Claude 3.7 Sonnet (Reasoning)ClosedAnthropic · 1 more variant	34.7	—	—	—
79	▼ 2	Gemini 2.5 ProClosedGoogle	34.6	$3.4	137	22 s
80	▼ 2	Nova 2.0 Lite (high)ClosedAmazon · 3 more variants	34.5	$0.85	166	15 s
81	▼ 2	DeepSeek V3.1 Terminus (Reasoning)OpenDeepSeek · 1 more variant	33.9	$1.9	—	—
82	▼ 2	Ling-2.6-1TOpenInclusionAI	33.6	$0.85	—	—
83	▼ 2	Doubao Seed CodeClosedByteDance Seed	33.5	—	—	—
84	New	Gemini 3.1 Flash-LiteClosedGoogle	33.5	$0.56	279	5.7 s
85	▼ 2	gpt-oss-120b (high)OpenOpenAI · 1 more variant	33.3	$0.26	315	0.9 s
86	▼ 2	o4-mini (high)ClosedOpenAI	33.1	$1.9	142	25 s
87	▼ 2	DeepSeek V3.2 Exp (Reasoning)OpenDeepSeek · 1 more variant	32.9	$0.31	—	—
88	▼ 2	Mercury 2ClosedInception	32.8	$0.38	825	3.0 s
89	▼ 2	GLM-4.6 (Reasoning)OpenZ AI · 1 more variant	32.5	$0.96	30	3.6 s
90	▼ 2	Qwen3.5 9B (Reasoning)OpenAlibaba · 1 more variant	32.4	$0.11	60	2.4 s
91	▼ 2	K-EXAONE (Reasoning)OpenLG AI Research · 1 more variant	32.1	—	—	—
92	▼ 2	Grok 3 mini Reasoning (high)ClosedxAI	32.1	$0.35	88	0.6 s
93	▼ 2	Trinity Large ThinkingOpenArcee AI	31.9	$0.40	149	1.2 s
94	▼ 2	Qwen3 MaxClosedAlibaba · 1 more variant	31.4	$3.0	32	2.6 s
95	▼ 2	Gemma 4 26B A4B (Reasoning)OpenGoogle · 1 more variant	31.2	$0.20	—	—
96	▼ 2	Gemini 2.5 Flash Preview (Sep '25) (Reasoning)ClosedGoogle · 3 more variants	31.1	—	—	—
97	▼ 2	Kimi K2 0905OpenKimi · 1 more variant	30.9	$1.1	25	2.5 s
98	▼ 2	o1ClosedOpenAI · 1 more variant	30.7	$26	80	31 s
99	▼ 2	Gemini 2.5 Pro Preview (Mar' 25)ClosedGoogle · 1 more variant	30.3	—	—	—
100	▼ 2	EXAONE 4.5 33BOpenLG AI Research	30.2	—	—	—
101	▼ 2	GLM-4.7-Flash (Reasoning)OpenZ AI · 1 more variant	30.1	$0.15	79	1.6 s
102	▼ 2	Qwen3 235B A22B 2507 (Reasoning)OpenAlibaba · 2 more variants	29.5	$0.84	56	2.9 s
103	▼ 2	DeepSeek V3.2 SpecialeOpenDeepSeek	29.4	—	—	—
104	▼ 2	ERNIE 5.0 Thinking PreviewClosedBaidu	29.1	—	—	—
105	▼ 2	Grok Code Fast 1ClosedxAI	28.7	—	—	—
106	▼ 2	Nemotron Cascade 2 30B A3BOpenNVIDIA	28.4	—	—	—
107	▼ 2	Qwen3 Coder NextOpenAlibaba	28.3	$0.56	119	1.4 s
108	▼ 2	DeepSeek V3.1 (Non-reasoning)OpenDeepSeek · 1 more variant	28.1	$0.83	—	—
109	▼ 2	Nova 2.0 Omni (medium)ClosedAmazon · 2 more variants	28.0	$0.85	—	—
110	▼ 2	Mistral Small 4 (Reasoning)OpenMistral · 1 more variant	27.8	$0.26	184	0.7 s
111	▼ 2	Qwen3 VL 235B A22B (Reasoning)OpenAlibaba	27.6	$2.2	34	4.9 s
112	▼ 2	Magistral Medium 1.2ClosedMistral	27.1	$2.8	38	1.7 s
113	▼ 2	Qwen3.5 4B (Reasoning)OpenAlibaba · 1 more variant	27.1	$0.060	186	0.5 s
114	▼ 2	DeepSeek R1 0528 (May '25)OpenDeepSeek · 1 more variant	27.1	$2.1	—	—
115	▼ 2	Gemini 2.5 Flash (Reasoning)ClosedGoogle · 1 more variant	27.0	$0.85	217	18 s
116	▼ 2	GPT-5 nano (high)ClosedOpenAI · 2 more variants	26.8	$0.14	183	73 s
117	▼ 2	Qwen3 Next 80B A3B (Reasoning)OpenAlibaba	26.7	$1.9	134	2.3 s
118	▼ 2	GLM-4.5 (Reasoning)OpenZ AI	26.4	$1.0	50	3.4 s
119	▼ 2	GPT-4.1ClosedOpenAI	26.3	$3.5	97	1.1 s
120	▼ 2	Ling 2.6 FlashOpenInclusionAI	26.2	$0.15	—	—
121	▼ 2	Solar Pro 3ClosedUpstage	25.9	—	—	—
122	▼ 2	Qwen3.5 Omni FlashClosedAlibaba	25.9	$0.28	243	1.9 s
123	▼ 2	o3-miniClosedOpenAI · 1 more variant	25.9	$1.9	142	7.7 s
124	▼ 2	o1-proClosedOpenAI	25.8	$263	—	—
125	▼ 2	JT-MINIClosedChina Mobile	25.4	—	—	—
126	▼ 2	Grok 3ClosedxAI	25.2	$8.0	—	—
127	▼ 2	Seed-OSS-36B-InstructOpenByteDance Seed	25.2	$0.30	36	3.4 s
128	▼ 2	Qwen3 235B A22B 2507 InstructOpenAlibaba	25.0	$0.36	53	2.3 s
129	▼ 2	Qwen3 Coder 480B A35B InstructOpenAlibaba	24.8	$0.68	62	3.0 s
130	▼ 2	Qwen3 VL 32B (Reasoning)OpenAlibaba	24.7	$2.6	89	2.8 s
131	▼ 2	Sonar Reasoning ProClosedPerplexity	24.6	—	—	—
132	▼ 2	gpt-oss-20B (high)OpenOpenAI · 1 more variant	24.5	$0.088	240	0.7 s
133	▼ 2	MiniMax M1 80kOpenMiniMax	24.4	$0.96	—	—
134	▼ 2	NVIDIA Nemotron 3 Nano 30B A3B (Reasoning)OpenNVIDIA · 1 more variant	24.3	$0.096	142	1.8 s
135	▼ 2	LongCat Flash LiteOpenLongCat	23.9	$0.000	106	6.1 s
136	▼ 2	GLM-4.6V (Reasoning)OpenZ AI · 1 more variant	23.4	$0.45	70	4.5 s
137	▼ 2	GLM-4.5-AirOpenZ AI	23.2	$0.37	80	3.2 s
138	▼ 2	Mi:dm K 2.5 ProClosedKorea Telecom	23.1	—	—	—
139	▼ 2	GPT-4.1 miniClosedOpenAI	22.9	$0.70	77	0.8 s
140	▼ 2	Mistral Large 3OpenMistral	22.8	$0.75	50	1.1 s
141	▼ 2	Ring-1TOpenInclusionAI	22.8	—	—	—
142	▼ 2	Qwen3 30B A3B 2507 (Reasoning)OpenAlibaba · 2 more variants	22.4	$0.67	135	2.4 s
143	▼ 2	DeepSeek V3 0324OpenDeepSeek · 1 more variant	22.3	$1.2	—	—
144	▼ 2	INTELLECT-3OpenPrime Intellect	22.2	—	—	—
145	▼ 2	Devstral 2OpenMistral	22.0	$0.000	63	1.3 s
146	▼ 2	Solar Open 100B (Reasoning)OpenUpstage	21.7	—	—	—
147	▼ 2	Grok 3 Reasoning BetaClosedxAI	21.6	—	—	—
148	▼ 2	Gemini 2.5 Flash-Lite Preview (Sep '25) (Reasoning)ClosedGoogle · 1 more variant	21.6	$0.18	—	—
149	▼ 2	Nemotron 3 Nano Omni 30B A3B ReasoningOpenNVIDIA	21.4	$0.13	302	1.0 s
150	▼ 2	Mistral Medium 3.1ClosedMistral	21.3	$0.80	63	1.5 s
151	▼ 2	MiniMax M1 40kOpenMiniMax	20.9	—	—	—
152	▼ 2	Qwen3 VL 235B A22B InstructOpenAlibaba	20.8	$0.70	51	2.8 s
153	▼ 2	o1-miniClosedOpenAI	20.4	—	—	—
154	▼ 2	Qwen3 Next 80B A3B InstructOpenAlibaba	20.1	$0.88	138	2.3 s
155	▼ 2	Qwen3 Coder 30B A3B InstructOpenAlibaba	20.0	$0.35	87	2.7 s
156	▼ 2	GPT-4.5 (Preview)ClosedOpenAI	20.0	—	—	—
157	▼ 2	QwQ 32BOpenAlibaba	19.7	$0.75	29	2.1 s
158	▼ 2	Qwen3 VL 30B A3B (Reasoning)OpenAlibaba	19.7	$0.34	110	2.3 s
159	▼ 2	Gemini 2.0 Flash Thinking Experimental (Jan '25)ClosedGoogle · 1 more variant	19.6	—	—	—
160	▼ 2	Devstral Small 2OpenMistral	19.5	$0.000	50	1.1 s
161	▼ 2	Motif-2-12.7B-ReasoningClosedMotif Technologies	19.1	—	—	—
162	▼ 2	Ling-1TOpenInclusionAI	19.0	—	—	—
163	▼ 2	Nova PremierClosedAmazon	19.0	$5.0	34	2.9 s
164	▼ 2	Solar Pro 2 (Preview) (Reasoning)ClosedUpstage · 3 more variants	18.8	—	—	—
165	▼ 2	Magistral Medium 1ClosedMistral	18.8	—	—	—
166	▼ 2	Gemma 4 E4B (Reasoning)OpenGoogle · 1 more variant	18.8	—	—	—
167	▼ 2	Mistral Medium 3ClosedMistral	18.8	$0.80	40	1.4 s
168	▼ 2	Llama Nemotron Super 49B v1.5 (Reasoning)OpenNVIDIA · 1 more variant	18.7	$0.18	49	1.3 s
169	▼ 2	Claude 3.5 HaikuClosedAnthropic	18.7	$1.8	—	—
170	▼ 2	Devstral MediumClosedMistral	18.7	$0.80	79	1.4 s
171	▼ 2	GPT-4o (Aug '24)ClosedOpenAI · 4 more variants	18.6	$4.4	106	1.1 s
172	▼ 2	Gemini 2.0 Flash (Feb '25)ClosedGoogle · 1 more variant	18.5	$0.26	—	—
173	▼ 2	Llama 3.3 Nemotron Super 49B v1 (Reasoning)OpenNVIDIA · 1 more variant	18.5	—	—	—
174	▼ 2	Llama 4 MaverickOpenMeta	18.4	$0.48	105	1.0 s
175	▼ 2	Qwen3 4B 2507 (Reasoning)OpenAlibaba · 2 more variants	18.2	—	—	—
176	▼ 2	Magistral Small 1.2OpenMistral	18.2	$0.75	106	0.8 s
177	▼ 2	Sarvam 105B (high)OpenSarvam	18.2	$0.000	98	2.1 s
178	▼ 2	Gemini 2.0 Pro Experimental (Feb '25)ClosedGoogle	18.1	—	—	—
179	▼ 2	Devstral Small (May '25)OpenMistral · 1 more variant	18.0	—	—	—
180	▼ 2	Claude 3 OpusClosedAnthropic	18.0	$33	—	—
181	New	MiniCPM5-1B (Non-reasoning)OpenOpenBMB	17.9	—	—	—
182	▼ 3	Sonar ReasoningClosedPerplexity	17.9	—	—	—
183	▼ 3	Gemini 2.5 Flash-Lite (Reasoning)ClosedGoogle · 1 more variant	17.6	$0.18	223	26 s
184	▼ 3	Llama 3.1 Instruct 405BOpenMeta	17.4	$3.7	36	2.3 s
185	▼ 3	Qwen3 VL 32B InstructOpenAlibaba	17.2	$1.2	73	2.7 s
186	▼ 3	DeepSeek R1 Distill Qwen 32BOpenDeepSeek	17.2	—	—	—
187	▼ 3	Magistral Small 1OpenMistral	16.8	—	—	—
188	▼ 3	EXAONE 4.0 32B (Reasoning)OpenLG AI Research · 1 more variant	16.7	—	—	—
189	▼ 3	Qwen3 VL 8B (Reasoning)OpenAlibaba	16.7	$0.66	112	2.4 s
190	▼ 3	Qwen3 32B (Reasoning)OpenAlibaba · 1 more variant	16.5	$0.28	86	2.6 s
191	▼ 3	DeepSeek R1 0528 Qwen3 8BOpenDeepSeek	16.4	—	—	—
192	▼ 3	Qwen3.5 2B (Reasoning)OpenAlibaba · 1 more variant	16.3	$0.040	—	—
193	▼ 3	Qwen2.5 MaxClosedAlibaba	16.3	$2.8	46	3.2 s
194	▼ 3	Qwen3 14B (Reasoning)OpenAlibaba · 1 more variant	16.2	$0.73	62	2.8 s
195	▼ 3	Nanbeige4.1-3BOpenNanbeige	16.1	—	—	—
196	▼ 3	Qwen3 VL 30B A3B InstructOpenAlibaba	16.0	$0.30	111	2.2 s
197	▼ 3	Gemini 1.5 Pro (Sep '24)ClosedGoogle · 1 more variant	16.0	—	—	—
198	▼ 3	Ministral 3 14BOpenMistral	16.0	$0.20	88	0.8 s
199	▼ 3	DeepSeek R1 Distill Llama 70BOpenDeepSeek	16.0	$0.79	45	1.7 s
200	▼ 3	Claude 3.5 Sonnet (Oct '24)ClosedAnthropic · 1 more variant	15.9	$6.6	—	—
201	▼ 3	DeepSeek R1 Distill Qwen 14BOpenDeepSeek	15.8	—	—	—
202	▼ 3	Falcon-H1R-7BOpenTII UAE	15.8	—	—	—
203	▼ 3	Ling-flash-2.0OpenInclusionAI	15.7	$0.25	73	2.7 s
204	▼ 3	Qwen3 Omni 30B A3B (Reasoning)OpenAlibaba	15.6	$0.43	92	2.0 s
205	▼ 3	Qwen2.5 Instruct 72BOpenAlibaba	15.6	$0.37	28	3.7 s
206	▼ 3	SonarClosedPerplexity	15.5	—	—	—
207	▼ 3	Step3 VL 10BOpenStepFun	15.5	—	—	—
208	▼ 3	Sonar ProClosedPerplexity	15.2	—	—	—
209	▼ 3	Gemma 4 E2B (Reasoning)OpenGoogle · 1 more variant	15.2	—	—	—
210	▼ 3	QwQ 32B-PreviewOpenAlibaba	15.2	—	—	—
211	▼ 3	GLM-4.5V (Reasoning)OpenZ AI · 1 more variant	15.1	$0.90	25	2.2 s
212	▼ 3	Mistral Large 2 (Nov '24)OpenMistral · 1 more variant	15.1	$3.0	31	1.7 s
213	▼ 3	Mistral Small 3.2OpenMistral	15.1	$0.13	124	0.6 s
214	▼ 3	Llama 3.1 Nemotron Ultra 253B v1 (Reasoning)OpenNVIDIA	15.0	$0.90	52	2.4 s
215	▼ 3	Qwen3 30B A3B 2507 InstructOpenAlibaba	15.0	$0.21	90	1.9 s
216	▼ 3	ERNIE 4.5 300B A47BOpenBaidu	15.0	$0.49	25	3.6 s
217	▼ 3	NVIDIA Nemotron Nano 12B v2 VL (Reasoning)OpenNVIDIA · 1 more variant	14.9	$0.30	—	—
218	▼ 3	Ministral 3 8BOpenMistral	14.8	$0.15	92	0.7 s
219	▼ 3	NVIDIA Nemotron Nano 9B V2 (Reasoning)OpenNVIDIA · 1 more variant	14.8	$0.070	122	0.7 s
220	▼ 3	Gemini 2.0 Flash-Lite (Feb '25)ClosedGoogle · 1 more variant	14.7	—	—	—
221	▼ 3	Granite 4.1 30BOpenIBM	14.7	—	—	—
222	▼ 3	NVIDIA Nemotron 3 Nano 4BOpenNVIDIA	14.7	—	—	—
223	▼ 3	Llama 3.3 Instruct 70BOpenMeta	14.5	$0.62	79	1.6 s
224	▼ 3	Mistral Small 3.1OpenMistral	14.5	$0.14	169	0.7 s
225	▼ 3	Llama 3.1 Nemotron Nano 4B v1.1 (Reasoning)OpenNVIDIA	14.4	—	—	—
226	▼ 3	Kimi Linear 48B A3B InstructOpenKimi	14.4	—	—	—
227	▼ 3	Qwen3 VL 8B InstructOpenAlibaba	14.3	$0.31	120	2.3 s
228	▼ 3	Ring-flash-2.0OpenInclusionAI	14.0	$0.25	—	—
229	▼ 3	Pixtral LargeOpenMistral	14.0	$3.0	53	1.6 s
230	▼ 3	Grok 2 (Dec '24)OpenxAI	13.9	—	—	—
231	▼ 3	Gemini 1.5 Flash (Sep '24)ClosedGoogle · 1 more variant	13.8	—	—	—
232	▼ 3	Qwen3 VL 4B (Reasoning)OpenAlibaba	13.7	—	—	—
233	▼ 3	GPT-4 TurboClosedOpenAI	13.7	$15	26	2.9 s
234	▼ 3	Llama 4 ScoutOpenMeta	13.5	$0.29	105	0.9 s
235	▼ 3	Nova ProClosedAmazon	13.5	$1.4	—	—
236	▼ 3	Llama 3.1 Nemotron Instruct 70BOpenNVIDIA	13.4	$1.2	293	0.5 s
237	▼ 3	Grok BetaClosedxAI	13.3	—	—	—
238	▼ 3	Qwen2.5 Instruct 32BOpenAlibaba	13.2	—	—	—
239	▼ 3	Qwen3 8B (Reasoning)OpenAlibaba · 1 more variant	13.2	$0.37	39	3.8 s
240	▼ 3	GPT-4.1 nanoClosedOpenAI	13.0	$0.18	155	0.5 s
241	▼ 3	Qwen3 4B 2507 InstructOpenAlibaba	12.9	—	—	—
242	▼ 3	Qwen2.5 Coder Instruct 32BOpenAlibaba	12.9	—	—	—
243	▼ 3	GPT-4ClosedOpenAI	12.8	$38	—	—
244	▼ 3	Mistral Small 3OpenMistral	12.7	$0.10	166	0.7 s
245	▼ 3	MiniCPM-V 4.6 1.3BOpenOpenBMB	12.7	—	—	—
246	▼ 3	Nova LiteClosedAmazon	12.7	$0.11	169	1.0 s
247	▼ 3	GPT-4o miniClosedOpenAI	12.6	$0.26	44	1.8 s
248	▼ 3	DeepSeek-V2.5 (Dec '24)OpenDeepSeek · 1 more variant	12.5	—	—	—
249	▼ 3	Llama 3.1 Instruct 70BOpenMeta	12.5	$0.56	36	1.9 s
250	▼ 3	Granite 4.1 8BOpenIBM	12.4	$0.063	108	0.8 s
251	▼ 3	Sarvam 30B (high)OpenSarvam	12.3	$0.000	166	1.9 s
252	▼ 3	Claude 3 HaikuClosedAnthropic	12.3	$0.50	—	—
253	▼ 3	Mistral SabaClosedMistral	12.1	—	—	—
254	▼ 3	DeepSeek R1 Distill Llama 8BOpenDeepSeek	12.1	—	—	—
255	▼ 3	R1 1776OpenPerplexity	12.0	—	—	—
256	▼ 3	Qwen2.5 TurboClosedAlibaba	12.0	$0.088	63	2.5 s
257	▼ 3	Reka Flash (Sep '24)ClosedReka AI	12.0	$0.35	58	2.5 s
258	▼ 3	Llama 3.2 Instruct 90B (Vision)OpenMeta	11.9	$1.4	57	1.2 s
258	▼ 3	Solar MiniOpenUpstage	11.9	$0.15	—	—
260	▼ 3	Llama 3.1 Instruct 8BOpenMeta	11.8	$0.10	155	0.9 s
261	▼ 3	Grok-1OpenxAI	11.7	—	—	—
262	▼ 3	Qwen2 Instruct 72BOpenAlibaba	11.7	—	—	—
263	▼ 3	Ministral 3 3BOpenMistral	11.2	$0.10	187	0.5 s
264	▼ 3	Gemini 1.5 Flash-8BClosedGoogle	11.1	—	—	—
265	▼ 3	Jamba 1.7 LargeOpenAI21 Labs	10.9	$3.5	61	1.5 s
266	▼ 3	Granite 4.0 H SmallOpenIBM	10.8	$0.11	398	10 s
267	▼ 3	Jamba 1.5 LargeOpenAI21 Labs	10.7	$3.5	—	—
268	▼ 3	Qwen3 Omni 30B A3B InstructOpenAlibaba	10.7	$0.43	95	2.0 s
269	▼ 3	DeepSeek-Coder-V2OpenDeepSeek	10.6	—	—	—
270	▼ 3	Jamba 1.6 LargeOpenAI21 Labs	10.6	$3.5	61	1.6 s
271	▼ 3	Qwen3.5 0.8B (Reasoning)OpenAlibaba · 1 more variant	10.5	$0.020	—	—
272	▼ 3	LFM2 24B A2BOpenLiquid AI	10.5	$0.053	120	0.6 s
273	▼ 3	Phi-4OpenMicrosoft	10.4	$0.22	40	2.1 s
274	▼ 3	Gemma 3 27B InstructOpenGoogle	10.3	$0.14	—	—
275	▼ 3	Claude 3 SonnetClosedAnthropic	10.3	$6.0	—	—
276	▼ 3	Nova MicroClosedAmazon	10.3	$0.061	296	1.0 s
277	▼ 3	Mistral Small (Sep '24)OpenMistral · 1 more variant	10.2	$0.30	166	0.7 s
278	▼ 3	Gemini 1.0 UltraClosedGoogle	10.1	—	—	—
279	▼ 3	Phi-3 Mini Instruct 3.8BOpenMicrosoft	10.1	—	—	—
280	▼ 3	Gemma 3n E4B Instruct Preview (May '25)OpenGoogle	10.1	—	—	—
281	▼ 3	Phi-4 Multimodal InstructOpenMicrosoft	10.0	$0.000	16	1.8 s
282	▼ 3	Qwen2.5 Coder Instruct 7BOpenAlibaba	10.0	—	—	—
283	▼ 3	Mistral Large (Feb '24)ClosedMistral	9.9	$6.0	—	—
284	▼ 3	Mixtral 8x22B InstructOpenMistral	9.8	—	—	—
285	▼ 3	Llama 2 Chat 7BOpenMeta	9.7	$0.10	—	—
286	▼ 3	Llama 3.2 Instruct 3BOpenMeta	9.7	$0.15	52	1.1 s
287	▼ 3	Jamba Reasoning 3BOpenAI21 Labs	9.6	—	—	—
288	▼ 3	Qwen3 VL 4B InstructOpenAlibaba	9.6	—	—	—
289	▼ 3	Qwen1.5 Chat 110BOpenAlibaba	9.5	—	—	—
290	▼ 3	Reka Flash 3OpenReka AI	9.5	$0.35	—	—
291	▼ 3	Claude 2.1ClosedAnthropic	9.3	—	—	—
292	▼ 3	Ling-mini-2.0OpenInclusionAI	9.2	—	—	—
293	▼ 3	DeepSeek R1 Distill Qwen 1.5BOpenDeepSeek	9.1	—	—	—
294	▼ 3	Claude 2.0ClosedAnthropic	9.1	—	—	—
294	▼ 3	DeepSeek-V2-ChatOpenDeepSeek	9.1	—	—	—
296	▼ 3	Mistral MediumClosedMistral	9.0	$4.1	59	1.7 s
297	▼ 3	GPT-3.5 TurboClosedOpenAI	9.0	$0.75	—	—
298	▼ 3	Llama 3 Instruct 70BOpenMeta	8.9	$1.2	—	—
299	▼ 3	Arctic InstructOpenSnowflake	8.8	—	—	—
299	▼ 3	Qwen Chat 72BOpenAlibaba	8.8	—	—	—
301	▼ 3	Gemma 3 12B InstructOpenGoogle	8.8	$0.14	—	—
302	▼ 3	LFM 40BClosedLiquid AI	8.8	—	—	—
303	▼ 3	Llama 3.2 Instruct 11B (Vision)OpenMeta	8.7	$0.25	51	0.7 s
304	▼ 3	PALM-2ClosedGoogle	8.6	—	—	—
305	▼ 3	Granite 4.1 3BOpenIBM	8.5	—	—	—
306	▼ 3	Gemini 1.0 ProClosedGoogle	8.5	—	—	—
307	▼ 3	DeepSeek Coder V2 Lite InstructOpenDeepSeek	8.5	—	—	—
308	▼ 3	Sarvam M (Reasoning)OpenSarvam	8.4	$0.000	—	—
309	▼ 3	Phi-4 Mini InstructOpenMicrosoft	8.4	$0.000	—	—
310	▼ 3	DeepSeek LLM 67B Chat (V1)OpenDeepSeek	8.4	—	—	—
310	▼ 3	Llama 2 Chat 70BOpenMeta	8.4	—	—	—
312	▼ 3	Llama 2 Chat 13BOpenMeta	8.4	—	—	—
313	▼ 3	Command-R+ (Apr '24)OpenCohere · 1 more variant	8.3	$6.0	—	—
314	▼ 3	OpenChat 3.5 (1210)OpenOpenChat	8.3	—	—	—
315	▼ 3	DBRX InstructOpenDatabricks	8.3	—	—	—
316	▼ 3	Exaone 4.0 1.2B (Reasoning)OpenLG AI Research · 1 more variant	8.3	—	—	—
317	▼ 3	LFM2.5-1.2B-ThinkingOpenLiquid AI	8.1	—	—	—
318	▼ 3	Jamba 1.7 MiniOpenAI21 Labs	8.1	—	—	—
319	▼ 3	LFM2 2.6BOpenLiquid AI	8.0	$0.000	—	—
320	▼ 3	LFM2.5-1.2B-InstructOpenLiquid AI	8.0	$0.000	—	—
321	▼ 3	Jamba 1.5 MiniOpenAI21 Labs	8.0	$0.25	—	—
322	▼ 3	Granite 4.0 H 1BOpenIBM	8.0	—	—	—
323	▼ 3	Qwen3 1.7B (Reasoning)OpenAlibaba · 1 more variant	8.0	$0.40	127	2.0 s
324	▼ 3	Jamba 1.6 MiniOpenAI21 Labs	7.9	$0.25	182	1.0 s
325	▼ 3	Mixtral 8x7B InstructOpenMistral	7.7	$0.51	—	—
326	▼ 3	Gemma 3 270MOpenGoogle	7.7	—	—	—
327	▼ 3	Apertus 70B InstructOpenSwiss AI Initiative	7.7	$1.3	—	—
328	▼ 3	Granite 4.0 MicroOpenIBM	7.7	—	—	—
329	▼ 3	Claude InstantClosedAnthropic	7.4	—	—	—
329	▼ 3	Llama 65BOpenMeta	7.4	—	—	—
329	▼ 3	Mistral 7B InstructOpenMistral	7.4	$0.21	82	0.7 s
329	▼ 3	Qwen Chat 14BOpenAlibaba	7.4	—	—	—
333	▼ 3	Granite 4.0 1BOpenIBM	7.3	—	—	—
334	▼ 3	LFM2 8B A1BOpenLiquid AI	7.0	$0.000	—	—
335	▼ 3	Granite 3.3 8B (Non-reasoning)OpenIBM	7.0	$0.085	339	27 s
336	▼ 3	Qwen3 0.6B (Reasoning)OpenAlibaba · 1 more variant	6.5	$0.40	192	1.7 s
337	▼ 3	Llama 3 Instruct 8BOpenMeta	6.4	$0.070	—	—
338	▼ 3	Gemma 3n E4B InstructOpenGoogle	6.4	$0.025	55	1.4 s
339	▼ 3	LFM2 1.2BOpenLiquid AI	6.3	$0.000	—	—
340	▼ 3	Gemma 3 4B InstructOpenGoogle	6.3	$0.050	—	—
341	▼ 3	Llama 3.2 Instruct 1BOpenMeta	6.3	$0.050	87	1.0 s
342	▼ 3	LFM2.5-VL-1.6BOpenLiquid AI	6.2	$0.000	—	—
343	▼ 3	Granite 4.0 350MOpenIBM	6.1	—	—	—
344	▼ 3	Apertus 8B InstructOpenSwiss AI Initiative	5.9	$0.13	—	—
345	▼ 3	Gemma 3 1B InstructOpenGoogle	5.6	$0.000	—	—
346	▼ 3	Granite 4.0 H 350MOpenIBM	5.4	—	—	—
347	▼ 3	Gemma 3n E2B InstructOpenGoogle	4.8	$0.000	—	—
348	▼ 3	Tiny Aya GlobalOpenCohere	4.7	$0.000	—	—

298 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