Model rankings

Rank	Change	Model	Intelligence	Price $/M	Speed tok/s	First token
1	New	GPT-5.5 (xhigh)ClosedOpenAI	60.2	$11	—	—
2	New	GPT-5.5 (high)ClosedOpenAI	58.9	$11	—	—
3	New	Claude Opus 4.7 (Adaptive Reasoning, Max Effort)ClosedAnthropic	57.3	$10	42	17 s
4	▼ 3	Gemini 3.1 Pro PreviewClosedGoogle	57.2	$4.5	130	29 s
5	▼ 3	GPT-5.4 (xhigh)ClosedOpenAI	56.8	$5.6	78	167 s
6	New	GPT-5.5 (medium)ClosedOpenAI	56.7	$11	—	—
7	New	Kimi K2.6OpenKimi	53.9	$1.7	100	1.1 s
8	New	MiMo-V2.5-ProClosedXiaomi	53.8	$1.5	61	2.8 s
9	▼ 6	GPT-5.3 Codex (xhigh)ClosedOpenAI	53.6	$4.8	73	88 s
10	▼ 6	Claude Opus 4.6 (Adaptive Reasoning, Max Effort)ClosedAnthropic	53.0	$10	41	8.6 s
11	▼ 6	Muse SparkClosedMeta	52.2	—	—	—
12	New	Claude Opus 4.7 (Non-reasoning, High Effort)ClosedAnthropic	51.8	$10	40	2.2 s
13	New	Qwen3.6 Max PreviewClosedAlibaba	51.8	$2.9	35	3.4 s
14	▼ 8	Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort)ClosedAnthropic	51.7	$6.0	48	127 s
15	▼ 8	GLM-5.1 (Reasoning)OpenZ AI	51.4	$2.2	49	1.5 s
16	▼ 8	GPT-5.2 (xhigh)ClosedOpenAI	51.3	$4.8	67	104 s
17	New	GPT-5.5 (low)ClosedOpenAI	50.8	$11	—	—
18	▼ 9	Qwen3.6 PlusClosedAlibaba	50.0	$1.1	53	3.0 s
19	▼ 9	GLM-5 (Reasoning)OpenZ AI	49.8	$1.6	70	1.7 s
20	▼ 9	Claude Opus 4.5 (Reasoning)ClosedAnthropic	49.7	$10	46	8.3 s
21	▼ 9	MiniMax-M2.7OpenMiniMax	49.6	$0.52	50	2.3 s
22	▼ 9	Grok 4.20 0309 v2 (Reasoning)ClosedxAI	49.3	$3.0	164	16 s
23	▼ 9	MiMo-V2-ProClosedXiaomi	49.2	—	—	—
24	▼ 9	GPT-5.2 Codex (xhigh)ClosedOpenAI	49.0	$4.8	93	69 s
25	▼ 9	GPT-5.4 mini (xhigh)ClosedOpenAI	48.9	$1.7	164	8.9 s
26	▼ 9	Grok 4.20 0309 (Reasoning)ClosedxAI	48.5	$3.0	133	16 s
27	▼ 9	Gemini 3 Pro Preview (high)ClosedGoogle	48.4	$4.5	134	28 s
28	▼ 9	GPT-5.1 (high)ClosedOpenAI	47.7	$3.4	121	31 s
29	▼ 9	Kimi K2.5 (Reasoning)OpenKimi	46.8	$1.2	38	2.8 s
30	▼ 9	GLM-5-TurboClosedZ AI	46.8	—	—	—
31	▼ 9	GPT-5.2 (medium)ClosedOpenAI	46.6	$4.8	—	—
32	New	DeepSeek V4 Flash (Reasoning, Max Effort)OpenDeepSeek	46.5	$0.18	85	1.0 s
33	▼ 10	Claude Opus 4.6 (Non-reasoning, High Effort)ClosedAnthropic	46.5	$10	38	1.8 s
34	▼ 10	Gemini 3 Flash Preview (Reasoning)ClosedGoogle	46.4	$1.1	169	7.1 s
35	New	Qwen3.6 27B (Reasoning)OpenAlibaba	45.8	$1.4	64	3.9 s
36	▼ 11	Qwen3.5 397B A17B (Reasoning)OpenAlibaba	45.1	$1.4	53	2.4 s
37	▼ 11	MiMo-V2-Omni-0327ClosedXiaomi	44.9	—	—	—
38	New	DeepSeek V4 Flash (Reasoning, High Effort)OpenDeepSeek	44.9	$0.18	—	—
39	▼ 12	GPT-5 (high)ClosedOpenAI	44.6	$3.4	86	67 s
39	▼ 12	GPT-5 Codex (high)ClosedOpenAI	44.6	$3.4	181	6.4 s
41	▼ 12	Claude Sonnet 4.6 (Non-reasoning, High Effort)ClosedAnthropic	44.4	$6.0	44	1.4 s
42	▼ 12	GPT-5.4 nano (xhigh)ClosedOpenAI	44.0	$0.46	154	3.7 s
43	▼ 12	GLM-5.1 (Non-reasoning)OpenZ AI	43.8	$2.2	45	2.1 s
44	New	Qwen3.6 35B A3B (Reasoning)OpenAlibaba	43.5	$0.84	209	2.5 s
45	▼ 13	MiMo-V2-OmniClosedXiaomi	43.4	—	—	—
46	▼ 13	GPT-5.1 Codex (high)ClosedOpenAI	43.1	$3.4	156	7.0 s
47	▼ 13	Claude Opus 4.5 (Non-reasoning)ClosedAnthropic	43.1	$10	46	1.4 s
48	▼ 13	Claude 4.5 Sonnet (Reasoning)ClosedAnthropic	43.0	$6.0	40	9.3 s
49	▼ 13	GLM 5V Turbo (Reasoning)ClosedZ AI	42.9	—	—	—
50	▼ 13	Claude Sonnet 4.6 (Non-reasoning, Low Effort)ClosedAnthropic	42.6	$6.0	46	1.3 s
51	▼ 13	GLM-4.7 (Reasoning)OpenZ AI	42.1	$1.0	103	1.2 s
52	▼ 13	Qwen3.5 27B (Reasoning)OpenAlibaba	42.1	$0.83	80	5.7 s
53	▼ 13	GPT-5 (medium)ClosedOpenAI	42.0	$3.4	76	37 s
54	▼ 13	Claude 4.1 Opus (Reasoning)ClosedAnthropic	42.0	$30	29	12 s
55	▼ 13	MiniMax-M2.5OpenMiniMax	41.9	$0.52	98	1.9 s
56	▼ 13	DeepSeek V3.2 (Reasoning)OpenDeepSeek	41.7	$0.32	31	2.1 s
57	▼ 13	Qwen3.5 122B A10B (Reasoning)OpenAlibaba	41.6	$1.1	143	2.5 s
58	▼ 13	Grok 4ClosedxAI	41.5	$6.0	59	12 s
59	▼ 13	MiMo-V2-Flash (Feb 2026)OpenXiaomi	41.5	$0.15	122	2.3 s
60	▼ 13	Gemini 3 Pro Preview (low)ClosedGoogle	41.3	$4.5	—	—
61	▼ 13	GPT-5 mini (high)ClosedOpenAI	41.2	$0.69	76	111 s
62	New	GPT-5.5 (Non-reasoning)ClosedOpenAI	40.9	$11	—	—
63	▼ 14	Kimi K2 ThinkingOpenKimi	40.9	$1.1	79	2.0 s
64	▼ 14	o3-proClosedOpenAI	40.7	$35	34	111 s
65	▼ 14	GLM-5 (Non-reasoning)OpenZ AI	40.6	$1.6	53	2.1 s
66	▼ 14	Qwen3.5 397B A17B (Non-reasoning)OpenAlibaba	40.1	$1.4	53	2.5 s
67	▼ 14	Qwen3 Max ThinkingClosedAlibaba	39.9	$2.4	38	3.8 s
68	▼ 14	MiniMax-M2.1OpenMiniMax	39.4	$0.52	86	2.1 s
69	▼ 14	MiMo-V2-Flash (Reasoning)OpenXiaomi	39.2	$0.15	123	2.2 s
70	▼ 14	GPT-5 (low)ClosedOpenAI	39.2	$3.4	76	10.0 s
71	▼ 14	Gemma 4 31B (Reasoning)OpenGoogle	39.2	$0.000	35	1.8 s
72	▼ 14	Claude 4 Opus (Reasoning)ClosedAnthropic	39.0	$30	29	8.8 s
73	▼ 14	GPT-5 mini (medium)ClosedOpenAI	38.9	$0.69	76	16 s
74	▼ 14	Claude 4 Sonnet (Reasoning)ClosedAnthropic	38.7	$6.0	39	11 s
75	▼ 14	GPT-5.1 Codex mini (high)ClosedOpenAI	38.6	$0.69	193	5.8 s
75	▼ 14	Qwen3.5 Omni PlusClosedAlibaba	38.6	$1.5	52	2.4 s
77	▼ 14	Grok 4.1 Fast (Reasoning)ClosedxAI	38.6	$0.28	135	6.9 s
78	▼ 14	Step 3.5 Flash 2603ClosedStepFun	38.5	$0.000	150	1.2 s
79	▼ 14	o3ClosedOpenAI	38.4	$3.5	93	9.3 s
80	▼ 14	GPT-5.4 nano (medium)ClosedOpenAI	38.1	$0.46	158	3.1 s
81	▼ 14	Step 3.5 FlashOpenStepFun	37.8	$0.15	140	1.1 s
82	▼ 14	GPT-5.4 mini (medium)ClosedOpenAI	37.7	$1.7	171	9.3 s
83	▼ 14	Kimi K2.5 (Non-reasoning)OpenKimi	37.3	$1.2	38	3.0 s
84	▼ 14	Qwen3.5 27B (Non-reasoning)OpenAlibaba	37.2	$0.83	85	5.9 s
85	▼ 14	Claude 4.5 Sonnet (Non-reasoning)ClosedAnthropic	37.1	$6.0	40	1.2 s
86	▼ 14	Qwen3.5 35B A3B (Reasoning)OpenAlibaba	37.1	$0.69	169	2.3 s
87	▼ 14	Claude 4.5 Haiku (Reasoning)ClosedAnthropic	37.1	$2.0	89	19 s
88	▼ 14	MiniMax-M2OpenMiniMax	36.1	$0.52	91	1.9 s
89	▼ 14	Claude 4.1 Opus (Non-reasoning)ClosedAnthropic	36.0	$30	29	1.5 s
90	▼ 14	NVIDIA Nemotron 3 Super 120B A12B (Reasoning)OpenNVIDIA	36.0	$0.41	154	1.3 s
91	▼ 14	Qwen3.5 122B A10B (Non-reasoning)OpenAlibaba	35.9	$1.1	143	2.5 s
92	▼ 14	Nova 2.0 Pro Preview (medium)ClosedAmazon	35.7	$3.4	116	10 s
93	▼ 14	GPT-5.4 (Non-reasoning)ClosedOpenAI	35.4	$5.6	59	0.9 s
94	▼ 14	Grok 4 Fast (Reasoning)ClosedxAI	35.1	$0.28	222	3.5 s
95	▼ 14	Gemini 3 Flash Preview (Non-reasoning)ClosedGoogle	35.1	$1.1	192	1.4 s
96	▼ 14	Claude 3.7 Sonnet (Reasoning)ClosedAnthropic	34.7	$6.0	—	—
97	▼ 14	Gemini 2.5 ProClosedGoogle	34.6	$3.4	127	21 s
98	▼ 14	Nova 2.0 Lite (high)ClosedAmazon	34.5	$0.85	130	17 s
99	▼ 14	GLM-4.7 (Non-reasoning)OpenZ AI	34.2	$1.0	99	1.2 s
100	▼ 14	DeepSeek V3.1 Terminus (Reasoning)OpenDeepSeek	33.9	$1.9	—	—
101	New	Ling-2.6-1TOpenInclusionAI	33.6	$0.85	68	3.1 s
102	▼ 15	GPT-5.2 (Non-reasoning)ClosedOpenAI	33.6	$4.8	56	0.9 s
103	▼ 15	Doubao Seed CodeClosedByteDance Seed	33.5	—	—	—
103	▼ 15	Gemini 3.1 Flash-Lite PreviewClosedGoogle	33.5	$0.56	329	5.2 s
105	▼ 15	gpt-oss-120B (high)OpenOpenAI	33.3	$0.26	211	0.9 s
106	▼ 15	o4-mini (high)ClosedOpenAI	33.1	$1.9	121	24 s
107	▼ 15	Claude 4 Opus (Non-reasoning)ClosedAnthropic	33.0	$30	30	1.6 s
107	▼ 15	Claude 4 Sonnet (Non-reasoning)ClosedAnthropic	33.0	$6.0	40	1.2 s
109	▼ 15	DeepSeek V3.2 Exp (Reasoning)OpenDeepSeek	32.9	$0.32	31	2.0 s
110	▼ 15	Mercury 2ClosedInception	32.8	$0.38	717	3.4 s
111	▼ 15	GLM-4.6 (Reasoning)OpenZ AI	32.5	$0.96	31	1.6 s
112	▼ 15	Qwen3 Max Thinking (Preview)ClosedAlibaba	32.5	$2.4	43	4.2 s
113	▼ 15	Qwen3.5 9B (Reasoning)OpenAlibaba	32.4	$0.11	45	1.0 s
114	▼ 15	Gemma 4 31B (Non-reasoning)OpenGoogle	32.3	—	—	—
115	▼ 15	K-EXAONE (Reasoning)OpenLG AI Research	32.1	—	—	—
116	▼ 15	DeepSeek V3.2 (Non-reasoning)OpenDeepSeek	32.1	$0.32	31	2.0 s
117	▼ 15	Grok 3 mini Reasoning (high)ClosedxAI	32.1	$0.35	208	0.6 s
118	▼ 15	Nova 2.0 Pro Preview (low)ClosedAmazon	31.9	$3.4	119	12 s
119	▼ 15	Trinity Large ThinkingOpenArcee AI	31.9	$0.40	116	1.1 s
120	New	Qwen3.6 35B A3B (Non-reasoning)OpenAlibaba	31.5	$0.84	190	2.6 s
121	▼ 16	Qwen3 MaxClosedAlibaba	31.4	$2.4	33	3.1 s
122	▼ 16	Gemma 4 26B A4B (Reasoning)OpenGoogle	31.2	$0.20	—	—
123	▼ 16	Gemini 2.5 Flash Preview (Sep '25) (Reasoning)ClosedGoogle	31.1	—	—	—
124	▼ 16	Claude 4.5 Haiku (Non-reasoning)ClosedAnthropic	31.1	$2.0	87	0.7 s
125	▼ 16	Kimi K2 0905OpenKimi	30.9	$1.1	18	3.2 s
126	▼ 16	Claude 3.7 Sonnet (Non-reasoning)ClosedAnthropic	30.8	$6.0	—	—
127	▼ 16	o1ClosedOpenAI	30.8	$26	89	33 s
128	▼ 16	Qwen3.5 35B A3B (Non-reasoning)OpenAlibaba	30.7	$0.69	154	2.1 s
129	▼ 16	MiMo-V2-Flash (Non-reasoning)OpenXiaomi	30.4	$0.15	120	2.4 s
130	▼ 16	Gemini 2.5 Pro Preview (Mar' 25)ClosedGoogle	30.3	—	—	—
131	▼ 16	GLM-4.6 (Non-reasoning)OpenZ AI	30.2	$1.0	32	1.6 s
132	▼ 16	GLM-4.7-Flash (Reasoning)OpenZ AI	30.2	$0.15	87	1.2 s
133	▼ 16	Nova 2.0 Lite (medium)ClosedAmazon	29.7	$0.85	158	19 s
134	▼ 16	Grok 4.20 0309 (Non-reasoning)ClosedxAI	29.7	$3.0	139	0.6 s
135	▼ 16	Gemini 2.5 Pro Preview (May' 25)ClosedGoogle	29.5	$3.4	—	—
136	▼ 16	Qwen3 235B A22B 2507 (Reasoning)OpenAlibaba	29.5	$2.6	44	3.1 s
137	▼ 16	DeepSeek V3.2 SpecialeOpenDeepSeek	29.4	—	—	—
138	▼ 16	ERNIE 5.0 Thinking PreviewClosedBaidu	29.1	—	—	—
139	▼ 16	Grok 4.20 0309 v2 (Non-reasoning)ClosedxAI	29.0	$3.0	126	0.7 s
140	▼ 16	Grok Code Fast 1ClosedxAI	28.7	$0.53	152	3.8 s
141	▼ 16	DeepSeek V3.1 Terminus (Non-reasoning)OpenDeepSeek	28.5	$0.45	—	—
142	▼ 16	DeepSeek V3.2 Exp (Non-reasoning)OpenDeepSeek	28.4	$0.32	31	2.1 s
143	▼ 16	Nemotron Cascade 2 30B A3BOpenNVIDIA	28.4	—	—	—
144	▼ 16	Qwen3 Coder NextOpenAlibaba	28.3	$0.56	149	1.1 s
145	▼ 16	DeepSeek V3.1 (Non-reasoning)OpenDeepSeek	28.1	$0.83	—	—
146	▼ 16	Nova 2.0 Omni (medium)ClosedAmazon	28.0	$0.85	—	—
147	▼ 16	Mistral Small 4 (Reasoning)OpenMistral	27.8	$0.26	147	0.8 s
148	▼ 16	DeepSeek V3.1 (Reasoning)OpenDeepSeek	27.7	$0.87	—	—
149	▼ 16	Qwen3 VL 235B A22B (Reasoning)OpenAlibaba	27.6	$2.6	42	3.1 s
150	▼ 16	GPT-5.1 (Non-reasoning)ClosedOpenAI	27.4	$3.4	100	1.3 s
151	▼ 16	Qwen3.5 9B (Non-reasoning)OpenAlibaba	27.3	—	—	—
152	▼ 16	Magistral Medium 1.2ClosedMistral	27.1	$2.8	66	1.7 s
153	▼ 16	Gemma 4 26B A4B (Non-reasoning)OpenGoogle	27.1	—	—	—
154	▼ 16	Qwen3.5 4B (Reasoning)OpenAlibaba	27.1	$0.060	174	0.9 s
155	▼ 16	DeepSeek R1 0528 (May '25)OpenDeepSeek	27.1	$2.1	—	—
156	▼ 16	Gemini 2.5 Flash (Reasoning)ClosedGoogle	27.0	$0.85	217	16 s
157	▼ 16	GPT-5 nano (high)ClosedOpenAI	26.8	$0.14	157	91 s
158	▼ 16	Qwen3 Next 80B A3B (Reasoning)OpenAlibaba	26.7	$1.9	148	2.2 s
159	▼ 16	GLM-4.5 (Reasoning)OpenZ AI	26.4	$1.0	45	3.2 s
160	▼ 16	Kimi K2OpenKimi	26.3	$1.0	34	1.8 s
161	▼ 16	GPT-4.1ClosedOpenAI	26.3	$3.5	97	1.0 s
162	New	Ling 2.6 FlashOpenInclusionAI	26.2	$0.15	207	1.1 s
163	▼ 17	Qwen3 Max (Preview)ClosedAlibaba	26.1	$2.4	45	4.3 s
164	▼ 17	GPT-5 nano (medium)ClosedOpenAI	25.9	$0.14	142	48 s
165	▼ 17	Qwen3.5 Omni FlashClosedAlibaba	25.9	$0.28	164	2.0 s
165	▼ 17	Solar Pro 3ClosedUpstage	25.9	—	—	—
167	▼ 17	o3-miniClosedOpenAI	25.9	$1.9	140	8.4 s
168	▼ 17	o1-proClosedOpenAI	25.8	$263	—	—
169	▼ 17	Gemini 2.5 Flash Preview (Sep '25) (Non-reasoning)ClosedGoogle	25.7	—	—	—
170	▼ 17	JT-MINIClosedChina Mobile	25.4	—	—	—
171	▼ 17	o3-mini (high)ClosedOpenAI	25.2	$1.9	131	30 s
172	▼ 17	Grok 3ClosedxAI	25.2	$6.0	76	1.4 s
173	▼ 17	Seed-OSS-36B-InstructOpenByteDance Seed	25.2	$0.30	36	3.1 s
174	▼ 17	Qwen3 235B A22B 2507 InstructOpenAlibaba	25.0	$1.2	54	2.5 s
175	▼ 17	Qwen3 Coder 480B A35B InstructOpenAlibaba	24.8	$3.0	53	3.2 s
176	▼ 17	Qwen3 VL 32B (Reasoning)OpenAlibaba	24.7	$2.6	89	2.6 s
177	▼ 17	Sonar Reasoning ProClosedPerplexity	24.6	—	—	—
178	▼ 17	Nova 2.0 Lite (low)ClosedAmazon	24.6	$0.85	142	14 s
179	▼ 17	gpt-oss-120B (low)OpenOpenAI	24.5	$0.26	212	0.9 s
179	▼ 17	gpt-oss-20B (high)OpenOpenAI	24.5	$0.10	229	0.7 s
181	▼ 17	MiniMax M1 80kOpenMiniMax	24.4	$0.96	—	—
182	▼ 17	GPT-5.4 nano (Non-Reasoning)ClosedOpenAI	24.4	$0.46	144	0.7 s
183	▼ 17	Gemini 2.5 Flash Preview (Reasoning)ClosedGoogle	24.3	—	—	—
184	▼ 17	NVIDIA Nemotron 3 Nano 30B A3B (Reasoning)OpenNVIDIA	24.3	$0.096	128	2.4 s
185	▼ 17	LongCat Flash LiteOpenLongCat	23.9	$0.000	113	7.2 s
186	▼ 17	GPT-5 (minimal)ClosedOpenAI	23.9	$3.4	63	1.6 s
187	▼ 17	o1-previewClosedOpenAI	23.7	$29	—	—
188	▼ 17	Grok 4.1 Fast (Non-reasoning)ClosedxAI	23.6	$0.28	129	0.6 s
189	▼ 17	GLM-4.6V (Reasoning)OpenZ AI	23.4	$0.45	25	2.6 s
190	▼ 17	K-EXAONE (Non-reasoning)OpenLG AI Research	23.4	—	—	—
191	▼ 17	GPT-5.4 mini (Non-Reasoning)ClosedOpenAI	23.3	$1.7	150	0.7 s
192	▼ 17	Nova 2.0 Omni (low)ClosedAmazon	23.2	$0.85	—	—
193	▼ 17	GLM-4.5-AirOpenZ AI	23.2	$0.37	82	3.0 s
194	▼ 17	Grok 4 Fast (Non-reasoning)ClosedxAI	23.1	$0.28	202	0.6 s
195	▼ 17	Mi:dm K 2.5 ProClosedKorea Telecom	23.1	—	—	—
195	▼ 17	Nova 2.0 Pro Preview (Non-reasoning)ClosedAmazon	23.1	$3.4	101	1.1 s
197	▼ 17	GPT-4.1 miniClosedOpenAI	22.9	$0.70	74	0.9 s
198	▼ 17	Mistral Large 3OpenMistral	22.8	$0.75	50	48 s
199	▼ 17	Ring-1TOpenInclusionAI	22.8	—	—	—
200	▼ 17	Qwen3.5 4B (Non-reasoning)OpenAlibaba	22.6	$0.060	171	0.7 s
201	▼ 17	Qwen3 30B A3B 2507 (Reasoning)OpenAlibaba	22.4	$0.75	144	2.5 s
202	▼ 17	DeepSeek V3 0324OpenDeepSeek	22.3	$1.2	—	—
203	▼ 17	INTELLECT-3OpenPrime Intellect	22.2	—	—	—
204	▼ 17	GLM-4.7-Flash (Non-reasoning)OpenZ AI	22.1	$0.15	83	1.7 s
205	▼ 17	Devstral 2OpenMistral	22.0	$0.000	72	1.1 s
206	▼ 17	GPT-5 (ChatGPT)ClosedOpenAI	21.8	$3.4	151	1.0 s
207	▼ 17	Solar Open 100B (Reasoning)OpenUpstage	21.7	—	—	—
208	▼ 17	Gemini 2.5 Flash-Lite Preview (Sep '25) (Reasoning)ClosedGoogle	21.7	$0.18	—	—
209	▼ 17	Grok 3 Reasoning BetaClosedxAI	21.6	—	—	—
210	▼ 17	Mistral Medium 3.1ClosedMistral	21.3	$0.80	47	1.3 s
211	▼ 17	MiniMax M1 40kOpenMiniMax	20.9	—	—	—
212	▼ 17	gpt-oss-20B (low)OpenOpenAI	20.8	$0.11	236	0.8 s
213	▼ 17	Qwen3 VL 235B A22B InstructOpenAlibaba	20.8	$1.2	51	2.7 s
214	▼ 17	GPT-5 mini (minimal)ClosedOpenAI	20.7	$0.69	78	1.3 s
215	▼ 17	Gemini 2.5 Flash (Non-reasoning)ClosedGoogle	20.6	$0.85	206	0.6 s
216	▼ 17	o1-miniClosedOpenAI	20.4	—	—	—
217	▼ 17	Qwen3 Next 80B A3B InstructOpenAlibaba	20.1	$0.88	149	2.3 s
218	▼ 17	Qwen3 Coder 30B A3B InstructOpenAlibaba	20.0	$0.90	107	2.6 s
219	▼ 17	GPT-4.5 (Preview)ClosedOpenAI	20.0	—	—	—
220	▼ 17	Qwen3 235B A22B (Reasoning)OpenAlibaba	19.8	$2.6	56	2.9 s
221	▼ 17	QwQ 32BOpenAlibaba	19.7	$0.75	29	2.1 s
222	▼ 17	Qwen3 VL 30B A3B (Reasoning)OpenAlibaba	19.7	$0.75	111	2.3 s
223	▼ 17	Gemini 2.0 Flash Thinking Experimental (Jan '25)ClosedGoogle	19.6	—	—	—
224	▼ 17	Devstral Small 2OpenMistral	19.5	$0.000	62	1.2 s
225	▼ 17	Gemini 2.5 Flash-Lite Preview (Sep '25) (Non-reasoning)ClosedGoogle	19.4	$0.18	—	—
226	▼ 17	Motif-2-12.7B-ReasoningClosedMotif Technologies	19.1	—	—	—
227	▼ 17	Ling-1TOpenInclusionAI	19.0	—	—	—
228	▼ 17	Nova PremierClosedAmazon	19.0	$5.0	26	3.2 s
229	▼ 17	DeepSeek R1 (Jan '25)OpenDeepSeek	18.8	$2.4	—	—
230	▼ 17	Solar Pro 2 (Preview) (Reasoning)ClosedUpstage	18.8	—	—	—
231	▼ 17	Magistral Medium 1ClosedMistral	18.8	—	—	—
232	▼ 17	Gemma 4 E4B (Reasoning)OpenGoogle	18.8	—	—	—
232	▼ 17	Mistral Medium 3ClosedMistral	18.8	$0.80	64	1.3 s
234	▼ 17	Llama Nemotron Super 49B v1.5 (Reasoning)OpenNVIDIA	18.7	$0.18	49	1.7 s
235	▼ 17	Claude 3.5 HaikuClosedAnthropic	18.7	$1.6	—	—
235	▼ 17	Devstral MediumClosedMistral	18.7	$0.80	133	1.2 s
237	▼ 17	GPT-4o (Aug '24)ClosedOpenAI	18.6	$4.4	63	1.4 s
238	▼ 17	Mistral Small 4 (Non-reasoning)OpenMistral	18.6	$0.26	136	0.7 s
239	▼ 17	GPT-4o (March 2025, chatgpt-4o-latest)ClosedOpenAI	18.6	—	—	—
240	▼ 17	Gemini 2.0 Flash (Feb '25)ClosedGoogle	18.5	$0.26	—	—
241	▼ 17	Llama 3.3 Nemotron Super 49B v1 (Reasoning)OpenNVIDIA	18.5	—	—	—
242	▼ 17	Llama 4 MaverickOpenMeta	18.4	$0.48	113	1.0 s
243	▼ 17	Qwen3 4B 2507 (Reasoning)OpenAlibaba	18.2	—	—	—
244	▼ 17	Magistral Small 1.2OpenMistral	18.2	$0.75	143	0.8 s
244	▼ 17	Sarvam 105B (high)OpenSarvam	18.2	$0.000	121	2.4 s
246	▼ 17	Gemini 2.0 Pro Experimental (Feb '25)ClosedGoogle	18.1	—	—	—
247	▼ 17	Devstral Small (May '25)OpenMistral	18.0	—	—	—
247	▼ 17	Nova 2.0 Lite (Non-reasoning)ClosedAmazon	18.0	$0.85	126	1.3 s
249	▼ 17	Claude 3 OpusClosedAnthropic	18.0	$30	—	—
250	▼ 17	Sonar ReasoningClosedPerplexity	17.9	—	—	—
251	▼ 17	Gemini 2.5 Flash Preview (Non-reasoning)ClosedGoogle	17.8	—	—	—
252	▼ 17	Gemini 2.5 Flash-Lite (Reasoning)ClosedGoogle	17.6	$0.18	305	19 s
253	▼ 17	Llama 3.1 Instruct 405BOpenMeta	17.4	$3.7	29	2.4 s
254	▼ 17	GPT-4o (Nov '24)ClosedOpenAI	17.3	$4.4	139	0.8 s
255	▼ 17	Qwen3 VL 32B InstructOpenAlibaba	17.2	$1.2	75	2.6 s
256	▼ 17	DeepSeek R1 Distill Qwen 32BOpenDeepSeek	17.2	—	—	—
257	▼ 17	GLM-4.6V (Non-reasoning)OpenZ AI	17.1	$0.45	23	6.1 s
258	▼ 17	Qwen3 235B A22B (Non-reasoning)OpenAlibaba	17.0	$1.2	56	2.8 s
259	▼ 17	Magistral Small 1OpenMistral	16.8	—	—	—
260	▼ 17	Gemini 2.0 Flash (experimental)ClosedGoogle	16.8	$0.000	—	—
261	▼ 17	EXAONE 4.0 32B (Reasoning)OpenLG AI Research	16.7	—	—	—
262	▼ 17	Qwen3 VL 8B (Reasoning)OpenAlibaba	16.7	$0.66	110	2.5 s
263	▼ 17	Nova 2.0 Omni (Non-reasoning)ClosedAmazon	16.6	$0.85	185	1.4 s
264	▼ 17	Qwen3 32B (Reasoning)OpenAlibaba	16.5	$2.6	89	2.5 s
265	▼ 17	DeepSeek V3 (Dec '24)OpenDeepSeek	16.5	$0.52	—	—
266	▼ 17	DeepSeek R1 0528 Qwen3 8BOpenDeepSeek	16.4	—	—	—
267	▼ 17	Qwen3.5 2B (Reasoning)OpenAlibaba	16.3	$0.040	—	—
268	▼ 17	Qwen2.5 MaxClosedAlibaba	16.3	$2.8	46	3.3 s
269	▼ 17	Qwen3 14B (Reasoning)OpenAlibaba	16.2	$1.3	62	2.8 s
270	▼ 17	Nanbeige4.1-3BOpenNanbeige	16.1	—	—	—
271	▼ 17	Qwen3 VL 30B A3B InstructOpenAlibaba	16.1	$0.35	111	2.2 s
272	▼ 17	Gemini 1.5 Pro (Sep '24)ClosedGoogle	16.0	—	—	—
272	▼ 17	Solar Pro 2 (Preview) (Non-reasoning)ClosedUpstage	16.0	—	—	—
274	▼ 17	Ministral 3 14BOpenMistral	16.0	$0.20	99	0.6 s
275	▼ 17	DeepSeek R1 Distill Llama 70BOpenDeepSeek	16.0	$0.79	44	1.8 s
276	▼ 17	Claude 3.5 Sonnet (Oct '24)ClosedAnthropic	15.9	$6.0	—	—
277	▼ 17	DeepSeek R1 Distill Qwen 14BOpenDeepSeek	15.8	—	—	—
278	▼ 17	Falcon-H1R-7BOpenTII UAE	15.8	—	—	—
279	▼ 17	Ling-flash-2.0OpenInclusionAI	15.7	$0.25	92	2.3 s
280	▼ 17	Qwen3 Omni 30B A3B (Reasoning)OpenAlibaba	15.6	$0.43	80	2.0 s
281	▼ 17	Qwen2.5 Instruct 72BOpenAlibaba	15.6	$0.000	28	3.7 s
282	▼ 17	SonarClosedPerplexity	15.5	—	—	—
283	▼ 17	Step3 VL 10BOpenStepFun	15.5	—	—	—
284	▼ 17	Qwen3 30B A3B (Reasoning)OpenAlibaba	15.3	$0.75	63	2.6 s
285	▼ 17	Sonar ProClosedPerplexity	15.2	—	—	—
286	▼ 17	Devstral Small (Jul '25)OpenMistral	15.2	$0.15	231	0.6 s
286	▼ 17	Gemma 4 E2B (Reasoning)OpenGoogle	15.2	—	—	—
288	▼ 17	QwQ 32B-PreviewOpenAlibaba	15.2	—	—	—
289	▼ 17	GLM-4.5V (Reasoning)OpenZ AI	15.1	$0.90	34	1.8 s
289	▼ 17	Mistral Large 2 (Nov '24)OpenMistral	15.1	$3.0	39	1.7 s
291	▼ 17	Mistral Small 3.2OpenMistral	15.1	$0.15	118	0.6 s
292	▼ 17	Llama 3.1 Nemotron Ultra 253B v1 (Reasoning)OpenNVIDIA	15.0	$0.90	42	2.6 s
293	▼ 17	Qwen3 30B A3B 2507 InstructOpenAlibaba	15.0	$0.35	55	2.2 s
294	▼ 17	ERNIE 4.5 300B A47BOpenBaidu	15.0	$0.49	26	3.7 s
295	▼ 17	Solar Pro 2 (Reasoning)ClosedUpstage	14.9	—	—	—
296	▼ 17	NVIDIA Nemotron Nano 12B v2 VL (Reasoning)OpenNVIDIA	14.9	$0.30	125	0.7 s
297	▼ 17	Ministral 3 8BOpenMistral	14.8	$0.15	176	0.6 s
298	▼ 17	Gemma 4 E4B (Non-reasoning)OpenGoogle	14.8	—	—	—
299	▼ 17	NVIDIA Nemotron Nano 9B V2 (Reasoning)OpenNVIDIA	14.8	$0.070	118	0.7 s
300	▼ 17	Gemini 2.0 Flash-Lite (Feb '25)ClosedGoogle	14.7	—	—	—
301	▼ 17	NVIDIA Nemotron 3 Nano 4BOpenNVIDIA	14.7	—	—	—
302	▼ 17	Qwen3.5 2B (Non-reasoning)OpenAlibaba	14.7	$0.040	223	0.5 s
303	▼ 17	Llama Nemotron Super 49B v1.5 (Non-reasoning)OpenNVIDIA	14.6	$0.18	49	1.7 s
304	▼ 17	Qwen3 32B (Non-reasoning)OpenAlibaba	14.5	$1.2	92	2.5 s
305	▼ 17	GPT-4o (May '24)ClosedOpenAI	14.5	$7.5	82	1.2 s
306	▼ 17	Llama 3.3 Instruct 70BOpenMeta	14.5	$0.62	91	1.4 s
307	▼ 17	Gemini 2.0 Flash-Lite (Preview)ClosedGoogle	14.5	—	—	—
308	▼ 17	Mistral Small 3.1OpenMistral	14.5	$0.15	128	0.7 s
309	▼ 17	Llama 3.1 Nemotron Nano 4B v1.1 (Reasoning)OpenNVIDIA	14.4	—	—	—
310	▼ 17	Kimi Linear 48B A3B InstructOpenKimi	14.4	—	—	—
311	▼ 17	Llama 3.3 Nemotron Super 49B v1 (Non-reasoning)OpenNVIDIA	14.3	—	—	—
312	▼ 17	Qwen3 VL 8B InstructOpenAlibaba	14.3	$0.31	120	2.4 s
313	▼ 17	Qwen3 4B (Reasoning)OpenAlibaba	14.2	$0.40	96	2.4 s
314	▼ 17	Claude 3.5 Sonnet (June '24)ClosedAnthropic	14.2	$6.0	—	—
315	▼ 17	GPT-4o (ChatGPT)ClosedOpenAI	14.1	—	—	—
316	▼ 17	Ring-flash-2.0OpenInclusionAI	14.0	$0.25	86	2.3 s
317	▼ 17	Pixtral LargeOpenMistral	14.0	$3.0	50	1.6 s
318	▼ 17	Grok 2 (Dec '24)OpenxAI	13.9	—	—	—
319	▼ 17	GPT-5 nano (minimal)ClosedOpenAI	13.8	$0.14	143	1.1 s
320	▼ 17	Gemini 1.5 Flash (Sep '24)ClosedGoogle	13.8	—	—	—
321	▼ 17	Qwen3 VL 4B (Reasoning)OpenAlibaba	13.7	—	—	—
322	▼ 17	GPT-4 TurboClosedOpenAI	13.7	$15	28	2.8 s
323	▼ 17	Solar Pro 2 (Non-reasoning)ClosedUpstage	13.6	—	—	—
324	▼ 17	Llama 4 ScoutOpenMeta	13.5	$0.29	139	0.8 s
325	▼ 17	Command AOpenCohere	13.5	$4.4	37	2.4 s
325	▼ 17	Nova ProClosedAmazon	13.5	$1.4	—	—
327	▼ 17	Llama 3.1 Nemotron Instruct 70BOpenNVIDIA	13.4	$1.2	39	1.7 s
328	▼ 17	Grok BetaClosedxAI	13.3	—	—	—
329	▼ 17	Qwen2.5 Instruct 32BOpenAlibaba	13.2	—	—	—
330	▼ 17	Qwen3 8B (Reasoning)OpenAlibaba	13.2	$0.66	84	2.3 s
331	▼ 17	NVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning)OpenNVIDIA	13.2	—	—	—
332	▼ 17	NVIDIA Nemotron Nano 9B V2 (Non-reasoning)OpenNVIDIA	13.2	$0.086	153	1.0 s
333	▼ 17	GPT-4.1 nanoClosedOpenAI	13.0	$0.18	165	0.7 s
334	▼ 17	Mistral Large 2 (Jul '24)OpenMistral	13.0	$3.0	—	—
335	▼ 17	Qwen3 4B 2507 InstructOpenAlibaba	12.9	—	—	—
336	▼ 17	Qwen2.5 Coder Instruct 32BOpenAlibaba	12.9	—	—	—
337	▼ 17	Qwen3 14B (Non-reasoning)OpenAlibaba	12.8	$0.61	62	2.8 s
338	▼ 17	GPT-4ClosedOpenAI	12.8	$38	—	—
339	▼ 17	GLM-4.5V (Non-reasoning)OpenZ AI	12.7	$0.90	42	22 s
340	▼ 17	Mistral Small 3OpenMistral	12.7	$0.15	132	0.8 s
341	▼ 17	Gemini 2.5 Flash-Lite (Non-reasoning)ClosedGoogle	12.7	$0.18	269	0.9 s
342	▼ 17	Nova LiteClosedAmazon	12.7	$0.11	165	1.0 s
343	▼ 17	GPT-4o miniClosedOpenAI	12.6	$0.26	54	1.2 s
344	▼ 17	Qwen3 30B A3B (Non-reasoning)OpenAlibaba	12.5	$0.35	62	2.4 s
345	▼ 17	DeepSeek-V2.5 (Dec '24)OpenDeepSeek	12.5	—	—	—
346	▼ 17	Qwen3 4B (Non-reasoning)OpenAlibaba	12.5	$0.19	94	2.3 s
347	▼ 17	Llama 3.1 Instruct 70BOpenMeta	12.5	$0.56	15	2.1 s
348	▼ 17	Sarvam 30B (high)OpenSarvam	12.3	$0.000	186	2.0 s
349	▼ 17	Gemini 2.0 Flash Thinking Experimental (Dec '24)ClosedGoogle	12.3	—	—	—
350	▼ 17	DeepSeek-V2.5OpenDeepSeek	12.3	—	—	—
351	▼ 17	Claude 3 HaikuClosedAnthropic	12.3	$0.50	—	—
352	▼ 17	Mistral SabaClosedMistral	12.1	—	—	—
353	▼ 17	DeepSeek R1 Distill Llama 8BOpenDeepSeek	12.1	—	—	—
354	▼ 17	Gemma 4 E2B (Non-reasoning)OpenGoogle	12.1	—	—	—
355	▼ 17	Gemini 1.5 Pro (May '24)ClosedGoogle	12.0	—	—	—
356	▼ 17	R1 1776OpenPerplexity	12.0	—	—	—
357	▼ 17	Qwen2.5 TurboClosedAlibaba	12.0	$0.088	63	2.4 s
358	▼ 17	Reka Flash (Sep '24)ClosedReka AI	12.0	$0.35	58	2.0 s
359	▼ 17	Llama 3.2 Instruct 90B (Vision)OpenMeta	11.9	$1.4	55	1.2 s
359	▼ 17	Solar MiniOpenUpstage	11.9	$0.15	—	—
361	▼ 17	Llama 3.1 Instruct 8BOpenMeta	11.8	$0.10	142	0.9 s
362	▼ 17	Grok-1OpenxAI	11.7	—	—	—
363	▼ 17	Qwen2 Instruct 72BOpenAlibaba	11.7	—	—	—
364	▼ 17	EXAONE 4.0 32B (Non-reasoning)OpenLG AI Research	11.7	—	—	—
365	▼ 17	Ministral 3 3BOpenMistral	11.2	$0.10	279	0.4 s
366	▼ 17	Gemini 1.5 Flash-8BClosedGoogle	11.1	—	—	—
367	▼ 17	Jamba 1.7 LargeOpenAI21 Labs	10.9	$3.5	62	1.5 s
368	▼ 17	Granite 4.0 H SmallOpenIBM	10.8	$0.11	430	10 s
369	▼ 17	Jamba 1.5 LargeOpenAI21 Labs	10.7	$3.5	—	—
370	▼ 17	Qwen3 Omni 30B A3B InstructOpenAlibaba	10.7	$0.43	95	1.9 s
371	▼ 17	Qwen3 8B (Non-reasoning)OpenAlibaba	10.6	$0.31	85	2.1 s
372	▼ 17	DeepSeek-Coder-V2OpenDeepSeek	10.6	—	—	—
373	▼ 17	Jamba 1.6 LargeOpenAI21 Labs	10.6	$3.5	62	1.4 s
374	▼ 17	Qwen3.5 0.8B (Reasoning)OpenAlibaba	10.5	$0.020	—	—
375	▼ 17	LFM2 24B A2BOpenLiquid AI	10.5	$0.053	122	0.5 s
376	▼ 17	Gemini 1.5 Flash (May '24)ClosedGoogle	10.5	—	—	—
377	▼ 17	Phi-4OpenMicrosoft Azure	10.4	$0.22	27	2.2 s
378	▼ 17	Gemma 3 27B InstructOpenGoogle	10.3	$0.000	28	2.0 s
379	▼ 17	Claude 3 SonnetClosedAnthropic	10.3	$6.0	—	—
380	▼ 17	Nova MicroClosedAmazon	10.3	$0.061	301	0.9 s
381	▼ 17	Mistral Small (Sep '24)OpenMistral	10.2	$0.30	132	0.7 s
382	▼ 17	Gemini 1.0 UltraClosedGoogle	10.1	—	—	—
383	▼ 17	Phi-3 Mini Instruct 3.8BOpenMicrosoft Azure	10.1	—	—	—
384	▼ 17	NVIDIA Nemotron Nano 12B v2 VL (Non-reasoning)OpenNVIDIA	10.1	$0.30	170	1.0 s
385	▼ 17	Gemma 3n E4B Instruct Preview (May '25)OpenGoogle	10.1	—	—	—
386	▼ 17	Phi-4 Multimodal InstructOpenMicrosoft Azure	10.0	$0.000	17	0.8 s
387	▼ 17	Qwen2.5 Coder Instruct 7BOpenAlibaba	10.0	—	—	—
388	▼ 17	Qwen3.5 0.8B (Non-reasoning)OpenAlibaba	9.9	$0.020	277	0.5 s
389	▼ 17	Mistral Large (Feb '24)ClosedMistral	9.9	$6.0	—	—
390	▼ 17	Mixtral 8x22B InstructOpenMistral	9.8	—	—	—
391	▼ 17	Llama 2 Chat 7BOpenMeta	9.7	$0.10	—	—
392	▼ 17	Llama 3.2 Instruct 3BOpenMeta	9.7	$0.15	51	1.1 s
393	▼ 17	Jamba Reasoning 3BOpenAI21 Labs	9.6	—	—	—
394	▼ 17	Qwen3 VL 4B InstructOpenAlibaba	9.6	—	—	—
395	▼ 17	Qwen1.5 Chat 110BOpenAlibaba	9.5	—	—	—
396	▼ 17	Reka Flash 3OpenReka AI	9.5	$0.35	—	—
397	▼ 17	Claude 2.1ClosedAnthropic	9.3	—	—	—
398	▼ 17	Ling-mini-2.0OpenInclusionAI	9.2	—	—	—
399	▼ 17	DeepSeek R1 Distill Qwen 1.5BOpenDeepSeek	9.1	—	—	—
400	▼ 17	Claude 2.0ClosedAnthropic	9.1	—	—	—
400	▼ 17	DeepSeek-V2-ChatOpenDeepSeek	9.1	—	—	—
402	▼ 17	Mistral Small (Feb '24)ClosedMistral	9.0	$1.5	134	0.7 s
403	▼ 17	Mistral MediumClosedMistral	9.0	$4.1	41	1.4 s
404	▼ 17	GPT-3.5 TurboClosedOpenAI	9.0	$0.75	—	—
405	▼ 17	Llama 3 Instruct 70BOpenMeta	8.9	$1.2	—	—
406	▼ 17	Arctic InstructOpenSnowflake	8.8	—	—	—
406	▼ 17	Qwen Chat 72BOpenAlibaba	8.8	—	—	—
408	▼ 17	Gemma 3 12B InstructOpenGoogle	8.8	$0.000	27	3.3 s
409	▼ 17	LFM 40BClosedLiquid AI	8.8	—	—	—
410	▼ 17	Llama 3.2 Instruct 11B (Vision)OpenMeta	8.7	$0.25	50	0.7 s
411	▼ 17	PALM-2ClosedGoogle	8.6	—	—	—
412	▼ 17	Gemini 1.0 ProClosedGoogle	8.5	—	—	—
413	▼ 17	DeepSeek Coder V2 Lite InstructOpenDeepSeek	8.5	—	—	—
414	▼ 17	Phi-4 Mini InstructOpenMicrosoft Azure	8.4	$0.000	44	0.8 s
414	▼ 17	Sarvam M (Reasoning)OpenSarvam	8.4	—	—	—
416	▼ 17	DeepSeek LLM 67B Chat (V1)OpenDeepSeek	8.4	—	—	—
416	▼ 17	Llama 2 Chat 70BOpenMeta	8.4	—	—	—
418	▼ 17	Llama 2 Chat 13BOpenMeta	8.4	—	—	—
419	▼ 17	Command-R+ (Apr '24)OpenCohere	8.3	$6.0	—	—
420	▼ 17	OpenChat 3.5 (1210)OpenOpenChat	8.3	—	—	—
421	▼ 17	DBRX InstructOpenDatabricks	8.3	—	—	—
422	▼ 17	Exaone 4.0 1.2B (Reasoning)OpenLG AI Research	8.3	—	—	—
423	▼ 17	Exaone 4.0 1.2B (Non-reasoning)OpenLG AI Research	8.1	—	—	—
424	▼ 17	LFM2.5-1.2B-ThinkingOpenLiquid AI	8.1	—	—	—
425	▼ 17	Jamba 1.7 MiniOpenAI21 Labs	8.1	—	—	—
426	▼ 17	LFM2 2.6BOpenLiquid AI	8.0	$0.000	—	—
426	▼ 17	LFM2.5-1.2B-InstructOpenLiquid AI	8.0	$0.000	—	—
428	▼ 17	Jamba 1.5 MiniOpenAI21 Labs	8.0	$0.25	—	—
429	▼ 17	Granite 4.0 H 1BOpenIBM	8.0	—	—	—
430	▼ 17	Qwen3 1.7B (Reasoning)OpenAlibaba	8.0	$0.40	128	1.8 s
431	▼ 17	Jamba 1.6 MiniOpenAI21 Labs	7.9	$0.25	181	0.9 s
432	▼ 17	Mixtral 8x7B InstructOpenMistral	7.7	$0.51	—	—
433	▼ 17	Gemma 3 270MOpenGoogle	7.7	—	—	—
434	▼ 17	Apertus 70B InstructOpenSwiss AI Initiative	7.7	$1.3	—	—
435	▼ 17	Granite 4.0 MicroOpenIBM	7.7	—	—	—
436	▼ 17	Claude InstantClosedAnthropic	7.4	—	—	—
436	▼ 17	Command-R (Mar '24)OpenCohere	7.4	$0.75	—	—
436	▼ 17	Llama 65BOpenMeta	7.4	—	—	—
436	▼ 17	Mistral 7B InstructOpenMistral	7.4	$0.25	165	0.6 s
436	▼ 17	Qwen Chat 14BOpenAlibaba	7.4	—	—	—
441	▼ 17	Granite 4.0 1BOpenIBM	7.3	—	—	—
442	▼ 17	LFM2 8B A1BOpenLiquid AI	7.0	$0.000	—	—
443	▼ 17	Granite 3.3 8B (Non-reasoning)OpenIBM	7.0	$0.085	356	26 s
444	▼ 17	Qwen3 1.7B (Non-reasoning)OpenAlibaba	6.8	$0.19	127	1.8 s
445	▼ 17	Qwen3 0.6B (Reasoning)OpenAlibaba	6.5	$0.40	163	1.6 s
446	▼ 17	Gemma 3n E4B InstructOpenGoogle	6.4	$0.025	29	1.3 s
446	▼ 17	Llama 3 Instruct 8BOpenMeta	6.4	$0.070	—	—
448	▼ 17	LFM2 1.2BOpenLiquid AI	6.3	$0.000	—	—
449	▼ 17	Gemma 3 4B InstructOpenGoogle	6.3	$0.000	29	1.6 s
450	▼ 17	Llama 3.2 Instruct 1BOpenMeta	6.3	$0.050	90	0.9 s
451	▼ 17	LFM2.5-VL-1.6BOpenLiquid AI	6.2	$0.000	—	—
452	▼ 17	Granite 4.0 350MOpenIBM	6.1	—	—	—
453	▼ 17	Apertus 8B InstructOpenSwiss AI Initiative	5.9	$0.13	—	—
454	▼ 17	Qwen3 0.6B (Non-reasoning)OpenAlibaba	5.7	$0.19	161	1.7 s
455	▼ 17	Gemma 3 1B InstructOpenGoogle	5.6	$0.000	49	0.9 s
456	▼ 17	Granite 4.0 H 350MOpenIBM	5.4	—	—	—
457	▼ 17	Gemma 3n E2B InstructOpenGoogle	4.8	$0.000	—	—
458	▼ 17	Tiny Aya GlobalOpenCohere	4.7	—	—	—

408 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