大模型排行榜 · AI 模型榜单

一页聚合主流 AI 基准与榜单——arena.ai 竞技场 Elo、Artificial Analysis 智能指数与语音竞技场、OpenRouter 用量——覆盖对话、代码、TTS、语音识别、图片、视频与智能体。

排名	模型	厂商	智能指数 ?	输入 ?	输出	缓存 ?
1	Claude Opus 5.5 (Max, Default Fallback)	Anthropic	57.6	4.00	20.0	0.200
2	Claude Sonnet 5.5 (Max, Default Fallback)	Anthropic	56	2.00	10.0	0.100
3	Claude Opus 5.5 (Xhigh, Default Fallback)	Anthropic	56	4.00	20.0	0.200
4	Muse Wonder Full (Max)	Meta	55.8	—	—	—
5	Claude Opus 5.5 (High, Default Fallback)	Anthropic	53.6	4.00	20.0	0.200
6	Claude Fable 5.1 (Max, Default Fallback)	Anthropic	53.4	10.0	50.0	0.250
7	Claude Fable 5.1 (Xhigh, Default Fallback)	Anthropic	53.2	10.0	50.0	0.250
8	GPT-6 Astra (Max)	OpenAI	52.7	10.0	50.0	1.00
9	Gemini 4 Argon (High)	Google	52.6	—	—	—
10	GPT-6 Astra (Xhigh)	OpenAI	52.4	10.0	50.0	1.00
11	Muse Wonder (Max)	Meta	52.4	4.00	20.0	0.200
12	Claude Sonnet 5.5 (Xhigh, Default Fallback)	Anthropic	51.9	2.00	10.0	0.100
13	GPT-6.1 Sol (Max)	OpenAI	51.8	2.00	10.0	0.100
14	Claude Fable 5.1 (High, Default Fallback)	Anthropic	51.2	10.0	50.0	0.250
15	Claude Opus 5.5 (Medium, Default Fallback)	Anthropic	51.2	4.00	20.0	0.200
16	GPT-6.1 Sol (Xhigh)	OpenAI	51	2.00	10.0	0.100
17	GPT-6 Astra (High)	OpenAI	50.9	10.0	50.0	1.00
18	Claude Opus 5 (Max)	Anthropic	50.8	5.00	25.0	0.500
19	GPT-6.1 Sol (High)	OpenAI	50.2	2.00	10.0	0.100
20	Claude Opus 5 (Xhigh)	Anthropic	49.7	5.00	25.0	0.500
21	Claude Fable 5 (Max, Opus 4.8 Fallback)	Anthropic	49.6	10.0	50.0	1.00
22	GPT-6 Astra (Medium)	OpenAI	49.6	10.0	50.0	1.00
23	Claude Fable 5.1 (Medium, Default Fallback)	Anthropic	48.9	10.0	50.0	0.250
24	Claude Opus 5 (High)	Anthropic	48.1	5.00	25.0	0.500
25	Muse Spark 1.3 (Max)	Meta	48.1	1.25	4.25	0.150
26	GPT-6.1 Sol (Medium)	OpenAI	47.8	2.00	10.0	0.100
27	GPT-6 Sol (Max)	OpenAI	47.6	2.00	10.0	0.200
28	GPT-5.6 Sol (Max)	OpenAI	47	2.00	10.0	0.200
29	Claude Fable 5.1 (Low, Default Fallback)	Anthropic	46.8	10.0	50.0	0.250
30	Claude Sonnet 5.5 (High, Default Fallback)	Anthropic	46.8	2.00	10.0	0.100
31	Grok 4.7 (Xhigh)	SpaceXAI	46.4	2.00	6.00	0.500
32	MiMo-V2.6-Pro	Xiaomi	46.3	0.435	0.870	0.0036
33	Grok 4.7 (High)	SpaceXAI	46.3	2.00	6.00	0.500
34	GPT-6 Astra (Low)	OpenAI	45.8	10.0	50.0	1.00
35	Qwen3.8 Max (0902)	Alibaba	45.4	2.00	6.00	0.250
36	Muse Spark 1.3 (Xhigh)	Meta	45.1	1.25	4.25	0.150
37	GLM-5.3 (Max)	Z AI	44.8	0.079	4.80	0.059
38	Claude Opus 5 (Medium)	Anthropic	44.8	5.00	25.0	0.500
39	Grok 4.6 (High)	SpaceXAI	44.3	2.00	6.00	0.500
40	Grok 4.6 (Xhigh)	SpaceXAI	44.2	2.00	6.00	0.500
41	GPT-6 Sol (Xhigh)	OpenAI	44.2	2.00	10.0	0.200
42	GPT-5.6 Sol (Xhigh)	OpenAI	44	2.00	10.0	0.200
43	Step 5 Preview	StepFun	43.7	1.00	2.70	0.050
44	Kimi K3 (Max)	Kimi	43.6	0.730	13.5	0.550
45	Claude Haiku 5.5 (Max)	Anthropic	43.4	0.100	0.500	0.010
46	Grok 4.6 (Medium)	SpaceXAI	42.8	2.00	6.00	0.500
47	GPT-6 Sol (High)	OpenAI	42.4	2.00	10.0	0.200
48	Claude Opus 5.5 (Low, Default Fallback)	Anthropic	42.3	4.00	20.0	0.200
49	GPT-5.6 Sol (High)	OpenAI	42.3	2.00	10.0	0.200
50	Grok 4.7 (Low)	SpaceXAI	42.2	2.00	6.00	0.500
51	GPT-5.6 Terra (Max)	OpenAI	42.1	2.00	12.0	0.200
52	GPT-6.1 Sol (Low)	OpenAI	42.1	2.00	10.0	0.100
53	GLM-5.3-Flash	Z AI	41.8	0.026	0.900	0.011
54	Claude Opus 4.8 (Max)	Anthropic	41.8	5.00	25.0	0.500
55	Claude Haiku 5.5 (Xhigh)	Anthropic	41.2	0.100	0.500	0.010
56	ILing 3.1 Flash	InclusionAI	41.1	—	—	—
57	Gemini 3.8 Flash (High)	Google	40.9	0.750	3.75	0.075
58	Claude Sonnet 5.5 (Medium, Default Fallback)	Anthropic	40.8	2.00	10.0	0.100
59	Claude Opus 4.7 (Max)	Anthropic	40.7	5.00	25.0	0.500
60	Qwen3.8 Max	Alibaba	40.2	2.00	6.00	0.250
61	Qwen3.8 2.4T A95B	Alibaba	39.9	2.00	6.00	0.250
62	Gemini 3.8 Flash (Medium)	Google	39.8	0.750	3.75	0.075
63	GPT-6 Sol (Medium)	OpenAI	39.8	2.00	10.0	0.200
64	Qwen3.8-Flash-Next	Alibaba	39.8	—	—	—
65	Muse Spark 1.2 (Xhigh)	Meta	39.6	1.25	4.25	0.150
66	Gemini 3.7 Flash (Medium)	Google	39.6	0.750	3.75	0.075
67	DeepSeek V4.1 Flash (Max)	DeepSeek	39.5	0.045	0.891	0.0054
68	Claude Opus 5 (Low)	Anthropic	39.4	5.00	25.0	0.500
69	GPT-5.6 Sol (Medium)	OpenAI	39.2	2.00	10.0	0.200
70	Gemini 3.7 Flash (High)	Google	39.1	0.750	3.75	0.075
71	GPT-5.4 (Xhigh)	OpenAI	39	2.50	15.0	0.250
72	Grok 4.5 (High)	SpaceXAI	38.8	2.00	6.00	0.300
73	GPT-5.5 (Xhigh)	OpenAI	38.4	5.00	30.0	0.500
74	Mistral Large 4 Preview	Mistral	38.4	—	—	—
75	Claude Sonnet 5 (Max)	Anthropic	38.2	2.00	10.0	0.200
76	GPT-6 Luna (Max)	OpenAI	38.1	0.100	0.500	0.010
77	GPT-5.6 Terra (Xhigh)	OpenAI	38	2.00	12.0	0.200
78	MiMo-V2.6-Flash	Xiaomi	37.9	0.140	0.280	0.0028
79	Claude Haiku 5.5 (High)	Anthropic	37.8	0.100	0.500	0.010
80	GPT-5.6 Luna (Max)	OpenAI	37.3	0.200	1.20	0.020
81	GPT-5.5 (High)	OpenAI	37	5.00	30.0	0.500
82	Gemini 3.7 Flash (Low)	Google	36.9	0.750	3.75	0.075
83	DeepSeek V4 Pro 0813 (Max)	DeepSeek	36	0.660	1.98	0.022
84	Claude Sonnet 5.5 (Low, Default Fallback)	Anthropic	35.9	2.00	10.0	0.100
85	Grok 4.6 (Low)	SpaceXAI	35.1	2.00	6.00	0.500
86	DeepSeek V4 Flash Vision (Max)	DeepSeek	34.8	0.216	0.647	0.0069
87	GPT-6 Luna (Xhigh)	OpenAI	34.6	0.100	0.500	0.010
88	GPT-5.6 Luna (Xhigh)	OpenAI	34.6	0.200	1.20	0.020
89	Claude Haiku 5.5 (Medium)	Anthropic	34.5	0.100	0.500	0.010
90	Claude Sonnet 5 (Xhigh)	Anthropic	34.4	2.00	10.0	0.200
91	GLM-5.3 (Low)	Z AI	34.3	0.079	4.80	0.059
92	DeepSeek V4 Flash 0731 (Max)	DeepSeek	34.3	0.014	0.576	0.011
93	GPT-6 Sol (Low)	OpenAI	34.2	2.00	10.0	0.200
94	GPT-5.6 Terra (High)	OpenAI	34.2	2.00	12.0	0.200
95	Gemini 3.6 Flash (High)	Google	34	0.750	3.75	0.075
96	CJT-4.1 Flash 236B A21B (Reasoning)	China Mobile	33.9	—	—	—
97	GPT-5.5 (Medium)	OpenAI	33.8	5.00	30.0	0.500
98	Qwen3.8 27B (Xhigh)	Alibaba	33.7	0.040	2.30	0.020
99	Muse Spark 1.1 (Xhigh)	Meta	33.7	1.25	4.25	0.150
100	GLM-5.2 (Max)	Z AI	33.7	0.060	7.00	0.059
101	Gemini 3.5 Flash (Medium)	Google	33.6	1.50	9.00	0.150
102	MMotif 3	Motif Technologies	33.6	—	—	—
103	GPT-5.6 Sol (Low)	OpenAI	33.5	2.00	10.0	0.200
104	Gemini 3.8 Flash (Low)	Google	33.5	0.750	3.75	0.075
105	GPT-6 Luna (High)	OpenAI	32.9	0.100	0.500	0.010
106	Gemini 3.5 Flash (High)	Google	32.6	1.50	9.00	0.150
107	GPT-5.3 Codex (Xhigh)	OpenAI	32.5	1.75	14.0	0.175
108	MMotif 3 (Beta)	Motif Technologies	32.3	—	—	—
109	GPT-5.6 Luna (High)	OpenAI	32.1	0.200	1.20	0.020
110	Claude Opus 4.6 (Max)	Anthropic	31.9	5.00	25.0	0.500
111	Claude Sonnet 5 (High)	Anthropic	31.7	2.00	10.0	0.200
112	Muse Spark	Meta	31.3	—	—	—
113	Claude Opus 4.7 (Non-reasoning, High)	Anthropic	30.9	5.00	25.0	0.500
114	GPT-5.5 (Low)	OpenAI	30.7	5.00	30.0	0.500
115	IK2 Horizon 375B A23B	Institute of Foundation Models	30.5	—	—	—
116	GPT-5.2 (Xhigh)	OpenAI	30.4	1.75	14.0	0.175
117	DeepSeek V4 Pro 0424 (Max)	DeepSeek	30.4	0.660	1.98	0.022
118	GPT-5.6 Terra (Medium)	OpenAI	30.1	2.00	12.0	0.200
119	Kimi K3 (Low)	Kimi	30.1	0.730	13.5	0.550
120	DeepSeek V4 Pro 0424 (High)	DeepSeek	30.1	0.660	1.98	0.022
121	Claude Sonnet 4.6 (Max)	Anthropic	30.1	3.00	15.0	0.300
122	GPT-6 Luna (Medium)	OpenAI	29.9	0.100	0.500	0.010
123	Gemini 3.1 Pro Preview	Google	29.7	2.00	12.0	0.200
124	Claude Haiku 5.5 (Low)	Anthropic	29.5	0.100	0.500	0.010
125	Qwen3.7 Max	Alibaba	29.5	1.48	4.42	0.295
126	MiniMax-M3	MiniMax	29.2	0.300	1.20	0.060
127	Claude Opus 4.5 (Reasoning)	Anthropic	29.1	5.00	25.0	0.500
128	MiMo-V2-Pro	Xiaomi	28.6	—	—	—
129	GPT-5.2 Codex (Xhigh)	OpenAI	28.5	1.75	14.0	0.175
130	GPT-6 Sol (Non-reasoning)	OpenAI	28.5	2.00	10.0	0.200
131	Qwen3.6 Max Preview	Alibaba	28.4	—	—	—
132	GPT-5.6 Sol (Non-reasoning)	OpenAI	28.3	2.00	10.0	0.200
133	USolar Pro 4	Upstage	28.2	—	—	—
134	NNex-N2-Pro (Based on Qwen3.5-397B-A17B)	Nex AGI	28.2	—	—	—
135	Claude Sonnet 5 (Medium)	Anthropic	28.1	2.00	10.0	0.200
136	Gemini 3 Pro Preview (High)	Google	28	—	—	—
137	GLM-5 (Reasoning)	Z AI	27.9	0.600	1.92	0.120
138	GPT-5.4 (Low)	OpenAI	27.6	2.50	15.0	0.250
139	Qwen3.8 27B (Medium)	Alibaba	27.6	0.040	2.30	0.020
140	GPT-5.6 Terra (Low)	OpenAI	27.5	2.00	12.0	0.200
141	CJT-4.1 Flash 236B A21B (Non-reasoning)	China Mobile	27.3	—	—	—
142	Grok Build 0.1 0616	SpaceXAI	27.2	1.00	2.00	0.200
143	Kimi K2.6 (Reasoning)	Kimi	27	0.650	3.41	0.150
144	Qwen3.6 Plus	Alibaba	27	0.325	1.95	—
145	MQuasar 438B (Max, Based on GLM-5.2)	Multiverse Computing	26.7	—	—	—
146	GLM-5-Turbo	Z AI	26.6	1.20	4.00	0.240
147	GPT-5.2 (Medium)	OpenAI	26.5	1.75	14.0	0.175
148	AApodex 1.1	Apodex	26.4	—	—	—
149	Claude Opus 4.6 (Non-reasoning, High)	Anthropic	26.4	5.00	25.0	0.500
150	Gemini 3 Flash Preview (Reasoning)	Google	26.3	0.500	3.00	0.050
151	Qwen3.8 27B (Low)	Alibaba	26.2	0.040	2.30	0.020
152	GLM-5.1 (Reasoning)	Z AI	26.1	0.966	3.04	0.179
153	MiMo-V2.5-Pro (Reasoning)	Xiaomi	26	0.435	0.870	0.0036
154	GPT-5.5 Instant (June 2026)	OpenAI	26	5.00	30.0	0.500
155	Kimi K2.7 Code	Kimi	25.8	0.671	3.35	0.180
156	Inkling Small	Thinking Machines	25.7	0.450	1.20	0.100
157	Grok 4.20 0309 v2 (Reasoning)	SpaceXAI	25.7	—	—	—
158	IK2 Horizon MoVA 36B A4B	Institute of Foundation Models	25.3	—	—	—
159	Hy3	Tencent	25.3	0.132	0.528	0.033
160	Grok 4.20 0309 (Reasoning)	SpaceXAI	25.2	1.25	2.50	0.200
161	MiMo-V2.5	Xiaomi	25.2	0.140	0.280	0.0028
162	Qwen3.7 Plus	Alibaba	25.2	0.320	1.28	0.064
163	MiMo-V2-Omni-0327	Xiaomi	25.1	—	—	—
164	Inkling (Xhigh)	Thinking Machines	25	1.00	4.05	0.170
165	GPT-5.6 Luna (Medium)	OpenAI	25	0.200	1.20	0.020
166	GPT-5 Codex (High)	OpenAI	24.9	—	—	—
167	Grok 4.3 (High)	SpaceXAI	24.9	1.25	2.50	0.200
168	Grok 4.3 (Medium)	SpaceXAI	24.8	1.25	2.50	0.200
169	Claude Sonnet 4.6 (Non-reasoning, High)	Anthropic	24.7	3.00	15.0	0.300
170	GPT-5.1 (High)	OpenAI	24.7	1.25	10.0	0.125
171	DeepSeek V4.1 Flash (Non-reasoning)	DeepSeek	24.7	0.045	0.891	0.0054
172	USolar Open2 250B	Upstage	24.7	—	—	—
173	ILing-3.0-flash-VL	InclusionAI	24.6	0.021	0.062	0.0042
174	DeepSeek V4 Flash 0420 (High)	DeepSeek	24.4	0.014	0.576	0.011
175	Claude Sonnet 5 (Low)	Anthropic	24.3	2.00	10.0	0.200
176	Grok 4.3 (Low)	SpaceXAI	24.3	1.25	2.50	0.200
177	GLM-5.1 (Non-reasoning)	Z AI	24.2	0.966	3.04	0.179
178	DeepSeek V4 Flash 0420 (Max)	DeepSeek	24.2	0.014	0.576	0.011
179	USolar Mini 4	Upstage	24.1	—	—	—
180	GPT-5.4 mini (Xhigh)	OpenAI	24.1	0.750	4.50	0.075
181	MiMo-V2-Omni	Xiaomi	23.9	—	—	—
182	Gemini 3.5 Flash (Minimal)	Google	23.8	1.50	9.00	0.150
183	GPT-5.1 Codex (High)	OpenAI	23.7	1.25	10.0	0.125
184	Claude Opus 4.5 (Non-reasoning)	Anthropic	23.7	5.00	25.0	0.500
185	Kimi K2.6 (Non-reasoning)	Kimi	23.6	0.650	3.41	0.150
186	GLM 5V Turbo (Reasoning)	Z AI	23.5	1.20	4.00	0.240
187	Kimi K2.5 (Reasoning)	Kimi	23.5	0.500	2.50	0.150
188	Claude Sonnet 4.6 (Non-reasoning, Low)	Anthropic	23.3	3.00	15.0	0.300
189	GPT-5.5 (Non-reasoning)	OpenAI	23.2	5.00	30.0	0.500
190	Claude Sonnet 5 (Non-reasoning, High)	Anthropic	23.2	2.00	10.0	0.200
191	GPT-5 (High)	OpenAI	23	1.25	10.0	0.125
192	Qwen3.5 27B (Reasoning)	Alibaba	22.9	0.260	2.60	—
193	Nemotron 3 Ultra 550B A55B (Reasoning)	NVIDIA	22.9	0.600	2.40	0.120
194	GPT-5 (Medium)	OpenAI	22.9	1.25	10.0	0.125
195	MiniMax-M2.5	MiniMax	22.8	0.270	1.08	0.027
196	MiniMax-M2.7	MiniMax	22.8	0.300	1.20	0.060
197	Claude 4.1 Opus (Reasoning)	Anthropic	22.8	15.0	75.0	1.50
198	Hy3-preview (Reasoning)	Tencent	22.7	0.132	0.528	0.033
199	GPT-5.5 Instant (May 2026)	OpenAI	22.7	5.00	30.0	0.500
200	ILing-3.0-flash-Fin	InclusionAI	22.6	0.042	0.123	0.0084
201	Grok 4	SpaceXAI	22.5	—	—	—
202	MiMo-V2-Flash (Feb 2026)	Xiaomi	22.4	—	—	—
203	GLM-5.2 (Non-reasoning)	Z AI	22.4	0.060	7.00	0.059
204	Gemini 3 Pro Preview (Low)	Google	22.3	—	—	—
205	Gemini 3.5 Flash-Lite	Google	22.2	0.300	2.50	0.030
206	GLM-4.7 (Reasoning)	Z AI	22.2	0.600	2.20	0.110
207	Kimi K2 Thinking	Kimi	22	0.600	2.50	0.150
208	o3-pro	OpenAI	21.9	20.0	80.0	—
209	AG9v3-39A5B	AI9Stars	21.8	—	—	—
210	GLM-5 (Non-reasoning)	Z AI	21.8	0.600	1.92	0.120
211	KKAT Coder Pro V2	KwaiKAT	21.7	—	—	—
212	SA.X-K2	SK Telecom	21.5	—	—	—
213	GPT-6 Luna (Low)	OpenAI	21.5	0.100	0.500	0.010
214	DeepSeek V3.2 (Reasoning)	DeepSeek	21.5	0.259	0.800	0.140
215	Qwen3.5 397B A17B (Non-reasoning)	Alibaba	21.4	0.550	3.50	0.225
216	Qwen3.6 27B (Reasoning)	Alibaba	21.4	0.320	3.20	—
217	Qwen3 Max Thinking	Alibaba	21.3	—	—	—
218	GPT-5.6 Luna (Low)	OpenAI	21	0.200	1.20	0.020
219	MiniMax-M2.1	MiniMax	20.9	0.300	1.20	0.030
220	DeepSeek V4 Pro 0424 (Non-reasoning)	DeepSeek	20.8	0.660	1.98	0.022
221	GPT-5 (Low)	OpenAI	20.8	1.25	10.0	0.125
222	GPT-5.6 Terra (Non-reasoning)	OpenAI	20.8	2.00	12.0	0.200
223	MiMo-V2-Flash (Reasoning)	Xiaomi	20.8	—	—	—
224	Claude 4.5 Sonnet (Reasoning)	Anthropic	20.7	3.00	15.0	0.300
225	GPT-5.4 nano (Xhigh)	OpenAI	20.7	0.200	1.25	0.020
226	IK2 Horizon 7B	Institute of Foundation Models	20.6	—	—	—
227	Claude 4 Opus (Reasoning)	Anthropic	20.6	—	—	—
228	GPT-5 mini (Medium)	OpenAI	20.6	0.250	2.00	0.025
229	Qwen3.5 Omni Plus	Alibaba	20.4	—	—	—
230	Grok 4.1 Fast (Reasoning)	SpaceXAI	20.4	—	—	—
231	DeepSeek V4 Pro 0813 (Non-reasoning)	DeepSeek	20.4	0.660	1.98	0.022
232	GPT-5.1 Codex mini (High)	OpenAI	20.4	0.250	2.00	0.030
233	o3	OpenAI	20.2	2.00	8.00	0.500
234	Qwen3.8 27B (Non-reasoning)	Alibaba	20.2	0.040	2.30	0.020
235	ILing 3.0 Flash	InclusionAI	20.1	0.021	0.063	0.0042
236	GPT-5.4 nano (Medium)	OpenAI	20	0.200	1.25	0.020
237	Qwen3.6 27B (Non-reasoning)	Alibaba	19.8	0.320	3.20	—
238	LK-EXAONE 2.0 0803	LG AI Research	19.7	—	—	—
239	GPT-5.4 mini (Medium)	OpenAI	19.7	0.750	4.50	0.075
240	Step 3.7 Flash	StepFun	19.5	0.200	1.15	0.040
241	Qwen3.5 27B (Non-reasoning)	Alibaba	19.4	0.260	2.60	—
242	Kimi K2.5 (Non-reasoning)	Kimi	19.4	0.500	2.50	0.150
243	Qwen3.5 35B A3B (Reasoning)	Alibaba	19.3	0.080	0.750	0.040
244	Claude 4.5 Sonnet (Non-reasoning)	Anthropic	19.3	3.00	15.0	0.300
245	LLongCat 2.0	LongCat	19.1	0.300	1.20	0.0060
246	Claude 4 Sonnet (Reasoning)	Anthropic	18.9	3.00	15.0	0.300
247	DeepSeek V4 Flash 0420 (Non-reasoning)	DeepSeek	18.9	0.014	0.576	0.011
248	CJT-35B-Flash	China Mobile	18.7	—	—	—
249	Claude 4.1 Opus (Non-reasoning)	Anthropic	18.6	15.0	75.0	1.50
250	MiniMax-M2	MiniMax	18.6	0.300	1.20	—
251	KKAT-Coder-Pro V1	KwaiKAT	18.6	—	—	—
252	GPT-6 Luna (Non-reasoning)	OpenAI	18.5	0.100	0.500	0.010
253	GLM-4.6 (Reasoning)	Z AI	18.5	0.500	2.00	0.100
254	Qwen3.5 397B A17B (Reasoning)	Alibaba	18.4	0.550	3.50	0.225
255	MiMo-V2.5-Pro (Non-reasoning)	Xiaomi	18.3	0.435	0.870	0.0036
256	GPT-5.4 (Non-reasoning)	OpenAI	18.2	2.50	15.0	0.250
257	Qwen3.6 35B A3B (Reasoning)	Alibaba	18.2	0.100	0.950	0.100
258	Grok 4 Fast (Reasoning)	SpaceXAI	17.9	—	—	—
259	Gemini 3 Flash Preview (Non-reasoning)	Google	17.9	0.500	3.00	0.050
260	Qwen3.5 122B A10B (Non-reasoning)	Alibaba	17.7	0.260	2.08	—
261	Claude 3.7 Sonnet (Reasoning)	Anthropic	17.7	—	—	—
262	Muse Glimmer (High)	Meta	17.5	—	—	—
263	GLM-4.7 (Non-reasoning)	Z AI	17.4	0.600	2.20	0.110
264	Step 3.5 Flash 2603	StepFun	17	0.100	0.300	—
265	ILing-2.6-1T	InclusionAI	17	—	—	—
266	GPT-5.2 (Non-reasoning)	OpenAI	17	1.75	14.0	0.175
267	Hy3-preview (Non-reasoning)	Tencent	17	0.132	0.528	0.033
268	BDoubao Seed Code	ByteDance Seed	16.9	—	—	—
269	Claude 4.5 Haiku (Reasoning)	Anthropic	16.9	1.00	5.00	0.100
270	GPT-5 mini (High)	OpenAI	16.8	0.250	2.00	0.025
271	Gemma 4 26B A4B (Reasoning)	Google	16.7	0.076	0.255	0.043
272	o4-mini (High)	OpenAI	16.7	1.10	4.40	0.275
273	Claude 4 Opus (Non-reasoning)	Anthropic	16.6	—	—	—
274	DeepSeek V3.2 Exp (Reasoning)	DeepSeek	16.6	0.259	0.800	0.140
275	Step 3.5 Flash	StepFun	16.6	0.100	0.300	—
276	Claude 4 Sonnet (Non-reasoning)	Anthropic	16.6	3.00	15.0	0.300
277	IRing-2.6-1T	InclusionAI	16.6	—	—	—
278	Qwen3 Max Thinking (Preview)	Alibaba	16.3	—	—	—
279	Gemini 2.5 Pro	Google	16.1	1.25	10.0	0.125
280	DeepSeek V3.2 (Non-reasoning)	DeepSeek	16	0.259	0.800	0.140
281	MiMo-V2-Flash (Non-reasoning)	Xiaomi	16	—	—	—
282	Qwen3.5 122B A10B (Reasoning)	Alibaba	15.6	0.260	2.08	—
283	IK2 Horizon 3.7B	Institute of Foundation Models	15.6	—	—	—
284	Gemini 3.1 Flash-Lite	Google	15.6	0.250	1.50	0.025
285	Qwen3 Max	Alibaba	15.6	—	—	—
286	Gemini 2.5 Flash Preview (Sep '25) (Reasoning)	Google	15.5	0.300	2.50	0.030
287	GPT-5.6 Luna (Non-reasoning)	OpenAI	15.5	0.200	1.20	0.020
288	Claude 4.5 Haiku (Non-reasoning)	Anthropic	15.4	1.00	5.00	0.100
289	Kimi K2 0905	Kimi	15.3	0.600	2.50	0.150
290	Claude 3.7 Sonnet (Non-reasoning)	Anthropic	15.3	—	—	—
291	o1	OpenAI	15.2	15.0	60.0	7.50
292	Qwen3.6 35B A3B (Non-reasoning)	Alibaba	15.2	0.100	0.950	0.100
293	Qwen3.5 35B A3B (Non-reasoning)	Alibaba	15.1	0.080	0.750	0.040
294	Gemini 2.5 Pro Preview (Mar' 25)	Google	15	1.25	10.0	0.125
295	GLM-4.7-Flash (Reasoning)	Z AI	14.9	0.060	0.400	—
296	GLM-4.6 (Non-reasoning)	Z AI	14.9	0.500	2.00	0.100
297	DeepSeek V3.1 Terminus (Reasoning)	DeepSeek	14.8	0.270	1.00	—
298	Gemma 4 31B (Reasoning)	Google	14.7	0.090	0.340	0.050
299	Grok 4.20 0309 (Non-reasoning)	SpaceXAI	14.6	1.25	2.50	0.200
300	Grok 3 mini Reasoning (High)	SpaceXAI	14.6	—	—	—
301	Gemini 2.5 Pro Preview (May' 25)	Google	14.5	1.25	10.0	0.125
302	DeepSeek V3.2 Speciale	DeepSeek	14.5	—	—	—
303	LK-EXAONE (Reasoning)	LG AI Research	14.4	—	—	—
304	ERNIE 5.0 Thinking Preview	Baidu	14.3	—	—	—
305	Nova 2.0 Pro Preview (Medium)	Amazon	14.2	—	—	—
306	Grok 4.20 0309 v2 (Non-reasoning)	SpaceXAI	14.2	—	—	—
307	Mistral Medium 3.5	Mistral	14.2	1.50	7.50	—
308	Gemma 4 12B (Reasoning)	Google	14.2	—	—	—
309	Grok Code Fast 1	SpaceXAI	14.1	—	—	—
310	Grok 4.3 (Non-reasoning)	SpaceXAI	14	1.25	2.50	0.200
311	DeepSeek V3.2 Exp (Non-reasoning)	DeepSeek	13.9	0.259	0.800	0.140
312	Gemma 4 31B (Non-reasoning)	Google	13.9	0.090	0.340	0.050
313	DeepSeek V3.1 Terminus (Non-reasoning)	DeepSeek	13.9	0.270	1.00	—
314	IMercury 2	Inception	13.8	0.250	0.750	0.025
315	SApriel-v1.5-15B-Thinker	ServiceNow	13.8	—	—	—
316	DeepSeek V3.1 (Non-reasoning)	DeepSeek	13.7	—	—	—
317	Nova 2.0 Omni (Medium)	Amazon	13.6	—	—	—
318	DeepSeek V3.1 (Reasoning)	DeepSeek	13.5	—	—	—
319	Nova 2.0 Lite (High)	Amazon	13.4	—	—	—
320	SApriel-v1.6-15B-Thinker	ServiceNow	13.4	—	—	—
321	Qwen3 VL 235B A22B (Reasoning)	Alibaba	13.4	—	—	—
322	Qwen3.5 9B (Non-reasoning)	Alibaba	13.3	0.100	0.150	—
323	GPT-5.1 (Non-reasoning)	OpenAI	13.3	1.25	10.0	0.125
324	LEXAONE 4.5 33B (Reasoning)	LG AI Research	13.2	—	—	—
325	Gemini 2.5 Flash (Reasoning)	Google	13.1	0.300	2.50	0.030
326	CCommand A+	Cohere	13.1	—	—	—
327	Qwen3.5 4B (Reasoning)	Alibaba	13.1	—	—	—
328	Gemma 4 26B A4B (Non-reasoning)	Google	13.1	0.076	0.255	0.043
329	DeepSeek R1 0528 (May '25)	DeepSeek	13.1	0.500	2.15	0.350
330	GPT-5 nano (High)	OpenAI	13	0.050	0.400	0.0050
331	Nemotron 3.5 Lightning	NVIDIA	12.9	0.070	0.200	0.040
332	GLM-4.5 (Reasoning)	Z AI	12.8	0.600	2.20	0.110
333	Nemotron 3 Super 120B A12B (Reasoning)	NVIDIA	12.8	0.080	0.450	—
334	Nova 2.0 Pro Preview (Low)	Amazon	12.8	—	—	—
335	Granite 4.2 30B	IBM	12.7	—	—	—
336	GPT-4.1	OpenAI	12.7	2.00	8.00	0.500
337	Kimi K2	Kimi	12.7	0.600	2.50	0.150
338	Qwen3 235B A22B 2507 (Reasoning)	Alibaba	12.7	0.450	3.50	—
339	Qwen3 Max (Preview)	Alibaba	12.6	—	—	—
340	o3-mini	OpenAI	12.5	1.10	4.40	0.550
341	OMiniCPM5-2B	OpenBMB	12.5	—	—	—
342	Qwen3.5 Omni Flash	Alibaba	12.5	—	—	—
343	GPT-5 nano (Medium)	OpenAI	12.5	0.050	0.400	0.0050
344	Nova 2.0 Lite (Medium)	Amazon	12.5	—	—	—
345	Gemini 2.5 Flash Preview (Sep '25) (Non-reasoning)	Google	12.4	0.300	2.50	0.030
346	o1-pro	OpenAI	12.4	150	600	—
347	IMercury 2.5	Inception	12.3	0.040	0.150	0.0040
348	CJT-MINI	China Mobile	12.2	—	—	—
349	Grok 3	SpaceXAI	12.1	—	—	—
350	BSeed-OSS-36B-Instruct	ByteDance Seed	12.1	—	—	—
351	Qwen3 235B A22B 2507 Instruct	Alibaba	12	—	—	—
352	Qwen3 Coder 480B A35B Instruct	Alibaba	11.9	—	—	—
353	Qwen3 VL 32B (Reasoning)	Alibaba	11.9	—	—	—
354	Sonar Reasoning Pro	Perplexity	11.8	2.00	8.00	—
355	Nova 2.0 Lite (Low)	Amazon	11.8	—	—	—
356	Magistral Medium 1.2	Mistral	11.8	—	—	—
357	MHyperNova 60B 2605 (High, Based on gpt-oss-120b)	Multiverse Computing	11.7	—	—	—
358	Gemini 2.5 Flash Preview (Reasoning)	Google	11.7	0.300	2.50	0.030
359	Nemotron Cascade 2 30B A3B	NVIDIA	11.7	—	—	—
360	MiniMax M1 80k	MiniMax	11.7	0.550	2.20	—
361	GPT-5.4 nano (Non-reasoning)	OpenAI	11.7	0.200	1.25	0.020
362	gpt-oss-120b (High)	OpenAI	11.6	0.150	0.600	0.075
363	IK2 Think V2	Institute of Foundation Models	11.5	—	—	—
364	LLongCat Flash Lite	LongCat	11.5	—	—	—
365	o1-preview	OpenAI	11.4	15.0	60.0	7.50
366	DeepSeek R1 (Jan '25)	DeepSeek	11.4	0.500	2.15	0.350
367	NHyperCLOVA X SEED Think (32B)	Naver	11.4	—	—	—
368	GPT-5 (Minimal)	OpenAI	11.4	1.25	10.0	0.125
369	Mistral Small 4 (Reasoning)	Mistral	11.3	—	—	—
370	Grok 4.1 Fast (Non-reasoning)	SpaceXAI	11.3	—	—	—
371	Qwen3.5 9B (Reasoning)	Alibaba	11.2	0.100	0.150	—
372	LK-EXAONE (Non-reasoning)	LG AI Research	11.2	—	—	—
373	Qwen3 Next 80B A3B (Reasoning)	Alibaba	11.2	0.150	1.20	—
374	GLM-4.6V (Reasoning)	Z AI	11.2	0.300	0.900	0.055
375	Grok 4 Fast (Non-reasoning)	SpaceXAI	11.1	—	—	—
376	GLM-4.5-Air	Z AI	11.1	0.130	0.850	0.025
377	ILing 3.0 Tiny	InclusionAI	11.1	—	—	—
378	Granite 4.2 8B	IBM	11.1	0.060	0.250	0.015
379	GPT-5.4 mini (Non-reasoning)	OpenAI	11.1	0.750	4.50	0.075
380	Nova 2.0 Omni (Low)	Amazon	11.1	—	—	—
381	o3-mini (High)	OpenAI	11	1.10	4.40	0.550
382	KMi:dm K 2.5 Pro	Korea Telecom	11	—	—	—
383	IRing-1T	InclusionAI	10.9	—	—	—
384	AG9v3-3B	AI9Stars	10.8	—	—	—
385	Qwen3.5 4B (Non-reasoning)	Alibaba	10.8	—	—	—
386	ATrinity Large Thinking	Arcee AI	10.8	0.250	0.800	0.060
387	PINTELLECT-3 (Based on GLM-4.5-Air)	Prime Intellect	10.6	—	—	—
388	GLM-4.7-Flash (Non-reasoning)	Z AI	10.6	0.060	0.400	—
389	USolar Open 100B (Reasoning)	Upstage	10.4	—	—	—
390	Gemini 2.5 Flash-Lite Preview (Sep '25) (Reasoning)	Google	10.4	0.100	0.400	0.010
391	GPT-5 (ChatGPT)	OpenAI	10.4	1.25	10.0	0.125
392	Grok 3 Reasoning Beta	SpaceXAI	10.4	—	—	—
393	Nemotron 3 Nano Omni 30B A3B Reasoning	NVIDIA	10.3	—	—	—
394	GPT-4.1 mini	OpenAI	10.2	0.400	1.60	0.100
395	gpt-oss-120b (Low)	OpenAI	10.2	0.150	0.600	0.075
396	gpt-oss-20b (Low)	OpenAI	10	0.018	0.090	0.0090
397	MiniMax M1 40k	MiniMax	10	0.550	2.20	—
398	Llama 4 Maverick	Meta	10	0.188	0.652	0.050
399	Nova 2.0 Pro Preview (Non-reasoning)	Amazon	10	—	—	—
400	CNorth Mini Code	Cohere	9.9	—	—	—
401	Gemini 2.5 Flash (Non-reasoning)	Google	9.9	0.300	2.50	0.030
402	IK2-V2 (High)	Institute of Foundation Models	9.9	—	—	—
403	GPT-5 mini (Minimal)	OpenAI	9.9	0.250	2.00	0.025
404	Qwen3 VL 235B A22B Instruct	Alibaba	9.9	0.210	1.90	0.100
405	Qwen3 30B A3B 2507 (Reasoning)	Alibaba	9.8	0.120	0.500	—
406	o1-mini	OpenAI	9.8	—	—	—
407	DeepSeek V3 0324	DeepSeek	9.7	—	—	—
408	ILing 2.6 Flash	InclusionAI	9.7	—	—	—
409	Qwen3 Next 80B A3B Instruct	Alibaba	9.6	0.100	1.10	0.070
410	GPT-4.5 (Preview)	OpenAI	9.6	2.50	15.0	0.250
411	TTri-21B-think Preview	Trillion Labs	9.6	—	—	—
412	Qwen3 Coder 30B A3B Instruct	Alibaba	9.6	0.070	0.280	—
413	Qwen3 235B A22B (Reasoning)	Alibaba	9.5	0.450	3.50	—
414	DiffusionGemma 26B A4B	Google	9.5	—	—	—
415	Qwen3 VL 30B A3B (Reasoning)	Alibaba	9.5	0.290	1.00	—
416	QwQ 32B	Alibaba	9.5	—	—	—
417	Gemma 4 12B (Non-reasoning)	Google	9.4	—	—	—
418	Gemini 2.0 Flash Thinking Experimental (Jan '25)	Google	9.4	—	—	—
419	Mistral Large 3	Mistral	9.3	—	—	—
420	Gemini 2.5 Flash-Lite Preview (Sep '25) (Non-reasoning)	Google	9.3	0.100	0.400	0.010
421	Mistral Medium 3.1	Mistral	9.2	0.400	2.00	0.040
422	ILing-1T	InclusionAI	9.2	—	—	—
423	MMotif-2-12.7B-Reasoning	Motif Technologies	9.2	—	—	—
424	Nova Premier	Amazon	9.2	—	—	—
425	Qwen3 Coder Next	Alibaba	9.2	0.120	0.800	0.070
426	Granite 4.2 3B	IBM	9.1	—	—	—
427	Magistral Medium 1	Mistral	9.1	—	—	—
428	USolar Pro 2 (Preview) (Reasoning)	Upstage	9.1	—	—	—
429	Mistral Small 4 (Non-reasoning)	Mistral	9	—	—	—
430	gpt-oss-20b (High)	OpenAI	9	0.018	0.090	0.0090
431	GPT-4o (March 2025, chatgpt-4o-latest)	OpenAI	9	2.50	10.0	1.25
432	Mistral Medium 3	Mistral	9	0.400	2.00	0.040
433	TTri-21B-Think	Trillion Labs	9	—	—	—
434	Devstral Medium	Mistral	9	0.400	2.00	0.040
435	IK2-V2 (Medium)	Institute of Foundation Models	9	—	—	—
436	Llama Nemotron Super 49B v1.5 (Reasoning)	NVIDIA	9	—	—	—
437	Claude 3.5 Haiku	Anthropic	8.9	—	—	—
438	Llama 3.3 Nemotron Super 49B v1 (Reasoning)	NVIDIA	8.9	—	—	—
439	NVIDIA Nemotron 3 Nano 30B A3B (Reasoning)	NVIDIA	8.9	—	—	—
440	Gemini 2.0 Flash (Feb '25)	Google	8.9	—	—	—
441	Gemma 4 E4B (Reasoning)	Google	8.9	—	—	—
442	Qwen3 4B 2507 (Reasoning)	Alibaba	8.8	—	—	—
443	SSarvam 105B (High)	Sarvam	8.8	—	—	—
444	OMiniCPM5-1B (Reasoning)	OpenBMB	8.8	—	—	—
445	Sonar Reasoning	Perplexity	8.7	—	—	—
446	Gemini 2.5 Flash Preview (Non-reasoning)	Google	8.7	0.300	2.50	0.030
447	Nova 2.0 Lite (Non-reasoning)	Amazon	8.7	—	—	—
448	Claude 3 Opus	Anthropic	8.7	—	—	—
449	Gemini 2.0 Pro Experimental (Feb '25)	Google	8.7	—	—	—
450	Devstral Small (May '25)	Mistral	8.7	—	—	—
451	OMiniCPM5-1B (Non-reasoning)	OpenBMB	8.7	—	—	—
452	Devstral 2	Mistral	8.6	—	—	—
453	Magistral Small 1.2	Mistral	8.6	—	—	—
454	Qwen3 32B (Reasoning)	Alibaba	8.6	0.080	0.280	—
455	DeepSeek V3 (Dec '24)	DeepSeek	8.5	—	—	—
456	Gemini 2.5 Flash-Lite (Reasoning)	Google	8.5	0.100	0.400	0.010
457	GLM-4.6V (Non-reasoning)	Z AI	8.4	0.300	0.900	0.055
458	Qwen3 VL 32B Instruct	Alibaba	8.4	—	—	—
459	NNanbeige4.1-3B	Nanbeige	8.4	—	—	—
460	GPT-4o (Nov '24)	OpenAI	8.4	2.50	10.0	1.25
461	DeepSeek R1 Distill Qwen 32B	DeepSeek	8.4	—	—	—
462	LLFM2.5-2.6B	Liquid AI	8.4	—	—	—
463	Qwen3 235B A22B (Non-reasoning)	Alibaba	8.3	0.450	3.50	—
464	Magistral Small 1	Mistral	8.2	—	—	—
465	Nova 2.0 Omni (Non-reasoning)	Amazon	8.2	—	—	—
466	Mistral Small 3.2	Mistral	8.2	—	—	—
467	LEXAONE 4.0 32B (Reasoning)	LG AI Research	8.2	—	—	—
468	Qwen3 14B (Reasoning)	Alibaba	8.2	0.120	0.240	—
469	Qwen3 VL 8B (Reasoning)	Alibaba	8.2	—	—	—
470	Gemini 2.0 Flash (experimental)	Google	8.2	—	—	—
471	DeepSeek R1 0528 Qwen3 8B	DeepSeek	8.1	—	—	—
472	Llama 4 Scout	Meta	8.1	0.180	0.590	—
473	Qwen2.5 Max	Alibaba	8	—	—	—
474	Claude 3.5 Sonnet (Oct '24)	Anthropic	7.9	—	—	—
475	Gemini 1.5 Pro (Sep '24)	Google	7.9	—	—	—
476	USolar Pro 2 (Preview) (Non-reasoning)	Upstage	7.9	—	—	—
477	DeepSeek R1 Distill Llama 70B	DeepSeek	7.9	—	—	—
478	Qwen3 VL 30B A3B Instruct	Alibaba	7.9	0.150	0.600	—
479	NHermes 4 - Llama-3.1 70B (Reasoning)	Nous Research	7.9	—	—	—
480	GPT-4.1 nano	OpenAI	7.8	0.100	0.400	0.025
481	ILing-flash-2.0	InclusionAI	7.8	—	—	—
482	Gemma 4 E2B (Reasoning)	Google	7.8	—	—	—
483	DeepSeek R1 Distill Qwen 14B	DeepSeek	7.8	—	—	—
484	USolar Pro 3	Upstage	7.8	0.150	0.600	0.015
485	Qwen3 Omni 30B A3B (Reasoning)	Alibaba	7.8	—	—	—
486	TFalcon-H1R-7B	TII UAE	7.8	—	—	—
487	Sonar	Perplexity	7.7	1.00	1.00	—
488	GPT-4o (Aug '24)	OpenAI	7.7	2.50	10.0	1.25
489	Llama 3.3 Instruct 70B	Meta	7.7	0.590	0.790	0.295
490	Step3 VL 10B	StepFun	7.7	—	—	—
491	Qwen2.5 Instruct 72B	Alibaba	7.7	—	—	—
492	GLM-4.5V (Reasoning)	Z AI	7.6	0.600	1.80	0.110
493	Sonar Pro	Perplexity	7.6	3.00	15.0	—
494	Qwen3 30B A3B (Reasoning)	Alibaba	7.6	0.120	0.500	—
495	Mistral Large 2 (Nov '24)	Mistral	7.6	—	—	—
496	QwQ 32B-Preview	Alibaba	7.6	—	—	—
497	Devstral Small (Jul '25)	Mistral	7.6	—	—	—
498	Gemma 4 E4B (Non-reasoning)	Google	7.5	—	—	—
499	NVIDIA Nemotron Nano 12B v2 VL (Reasoning)	NVIDIA	7.5	—	—	—
500	Qwen3 30B A3B 2507 Instruct	Alibaba	7.5	0.100	0.300	—
501	NHermes 4 - Llama-3.1 405B (Reasoning)	Nous Research	7.5	—	—	—
502	ERNIE 4.5 300B A47B	Baidu	7.5	—	—	—
503	Devstral Small 2	Mistral	7.5	—	—	—
504	Llama 3.1 Nemotron Ultra 253B v1 (Reasoning)	NVIDIA	7.5	—	—	—
505	USolar Pro 2 (Reasoning)	Upstage	7.5	—	—	—
506	Llama Nemotron Super 49B v1.5 (Non-reasoning)	NVIDIA	7.4	—	—	—
507	Gemini 2.0 Flash-Lite (Feb '25)	Google	7.4	—	—	—
508	NHermes 4 - Llama-3.1 405B (Non-reasoning)	Nous Research	7.4	—	—	—
509	Granite 4.1 30B	IBM	7.4	—	—	—
510	NVIDIA Nemotron Nano 9B V2 (Reasoning)	NVIDIA	7.4	—	—	—
511	Qwen3 VL 8B Instruct	Alibaba	7.3	0.250	0.750	0.120
512	Gemini 2.0 Flash-Lite (Preview)	Google	7.3	—	—	—
513	Llama 3.1 Instruct 405B	Meta	7.3	—	—	—
514	Kimi Linear 48B A3B Instruct	Kimi	7.3	—	—	—
515	Qwen3 8B (Reasoning)	Alibaba	7.3	—	—	—
516	Llama 3.1 Nemotron Nano 4B v1.1 (Reasoning)	NVIDIA	7.3	—	—	—
517	Qwen3 32B (Non-reasoning)	Alibaba	7.3	0.080	0.280	—
518	Llama 3.3 Nemotron Super 49B v1 (Non-reasoning)	NVIDIA	7.3	—	—	—
519	IK2-V2 (Low)	Institute of Foundation Models	7.3	—	—	—
520	GPT-4o (May '24)	OpenAI	7.3	2.50	10.0	1.25
521	Qwen3 4B (Reasoning)	Alibaba	7.2	—	—	—
522	IRing-flash-2.0	InclusionAI	7.2	—	—	—
523	Claude 3.5 Sonnet (June '24)	Anthropic	7.2	—	—	—
524	GPT-4o (ChatGPT)	OpenAI	7.2	2.50	10.0	1.25
525	LLFM2.5-8B-A1B	Liquid AI	7.2	—	—	—
526	ALlama 3.1 Tulu3 405B	Allen Institute for AI	7.2	—	—	—
527	GPT-5 nano (Minimal)	OpenAI	7.1	0.050	0.400	0.0050
528	Mistral Small 3.1	Mistral	7.1	—	—	—
529	Grok 2 (Dec '24)	SpaceXAI	7.1	—	—	—
530	AOlmo 3.1 32B Think	Allen Institute for AI	7.1	—	—	—
531	Gemini 1.5 Flash (Sep '24)	Google	7.1	—	—	—
532	Pixtral Large	Mistral	7.1	—	—	—
533	CCommand A	Cohere	7	2.50	10.0	—
534	USolar Pro 2 (Non-reasoning)	Upstage	7	—	—	—
535	Nova Pro	Amazon	7	—	—	—
536	GPT-4 Turbo	OpenAI	7	10.0	30.0	—
537	Qwen3 VL 4B (Reasoning)	Alibaba	7	—	—	—
538	Qwen2.5 Instruct 32B	Alibaba	6.9	—	—	—
539	Qwen3.5 2B (Reasoning)	Alibaba	6.9	—	—	—
540	Llama 3.1 Nemotron Instruct 70B	NVIDIA	6.9	—	—	—
541	Llama 3.1 Instruct 8B	Meta	6.9	0.050	0.080	0.025
542	Grok Beta	SpaceXAI	6.9	2.00	6.00	0.500
543	Mistral Large 2 (Jul '24)	Mistral	6.8	—	—	—
544	NVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning)	NVIDIA	6.8	—	—	—
545	NVIDIA Nemotron Nano 9B V2 (Non-reasoning)	NVIDIA	6.8	—	—	—
546	GLM-4.5V (Non-reasoning)	Z AI	6.7	0.600	1.80	0.110
547	Qwen2.5 Coder Instruct 32B	Alibaba	6.7	—	—	—
548	Gemini 2.5 Flash-Lite (Non-reasoning)	Google	6.7	0.100	0.400	0.010
549	NHermes 4 - Llama-3.1 70B (Non-reasoning)	Nous Research	6.7	—	—	—
550	Nova Lite	Amazon	6.7	—	—	—
551	GPT-4	OpenAI	6.7	30.0	60.0	—
552	Qwen3 4B 2507 Instruct	Alibaba	6.7	—	—	—
553	GPT-4o mini	OpenAI	6.7	0.150	0.600	0.075
554	Mistral Small 3	Mistral	6.7	—	—	—
555	Qwen3 14B (Non-reasoning)	Alibaba	6.7	0.120	0.240	—
556	Llama 3.1 Instruct 70B	Meta	6.6	0.400	0.400	—
557	SSarvam 30B (High)	Sarvam	6.6	—	—	—
558	Granite 4.1 8B	IBM	6.6	—	—	—
559	DeepSeek-V2.5	DeepSeek	6.6	—	—	—
560	DeepSeek-V2.5 (Dec '24)	DeepSeek	6.6	—	—	—
561	Gemini 2.0 Flash Thinking Experimental (Dec '24)	Google	6.6	—	—	—
562	Qwen3 4B (Non-reasoning)	Alibaba	6.6	—	—	—
563	Qwen3 30B A3B (Non-reasoning)	Alibaba	6.6	0.120	0.500	—
564	DeepSeek R1 Distill Llama 8B	DeepSeek	6.5	—	—	—
565	Gemma 4 E2B (Non-reasoning)	Google	6.5	—	—	—
566	Mistral Saba	Mistral	6.5	0.200	0.600	0.020
567	AOlmo 3.1 32B Instruct	Allen Institute for AI	6.5	—	—	—
568	AOlmo 3 32B Think	Allen Institute for AI	6.5	—	—	—
569	Gemini 1.5 Pro (May '24)	Google	6.4	—	—	—
570	Qwen2.5 Turbo	Alibaba	6.4	—	—	—
571	R1 1776	Perplexity	6.4	—	—	—
572	Llama 3.2 Instruct 90B (Vision)	Meta	6.4	—	—	—
573	RReka Flash (Sep '24)	Reka AI	6.4	—	—	—
574	USolar Mini	Upstage	6.4	—	—	—
575	LEXAONE 4.0 32B (Non-reasoning)	LG AI Research	6.3	—	—	—
576	Grok-1	SpaceXAI	6.3	—	—	—
577	Qwen2 Instruct 72B	Alibaba	6.3	—	—	—
578	CCeleris-1	Celeris	6.3	—	—	—
579	Phi-4 Mini Instruct	Microsoft	6.3	—	—	—
580	Qwen3.5 2B (Non-reasoning)	Alibaba	6.2	—	—	—
581	Gemini 1.5 Flash-8B	Google	6.2	—	—	—
582	AJamba 1.7 Large	AI21 Labs	6.1	—	—	—
583	Qwen3.5 0.8B (Reasoning)	Alibaba	6.1	—	—	—
584	NDeepHermes 3 - Mistral 24B Preview (Non-reasoning)	Nous Research	6.1	—	—	—
585	Qwen3 Omni 30B A3B Instruct	Alibaba	6	—	—	—
586	NHermes 3 - Llama-3.1 70B	Nous Research	6	0.700	0.700	—
587	AOLMo 2 32B	Allen Institute for AI	6	—	—	—
588	Granite 4.0 H Small	IBM	6	—	—	—
589	Ministral 3 14B	Mistral	6	—	—	—
590	DeepSeek-Coder-V2	DeepSeek	6	—	—	—
591	AJamba 1.6 Large	AI21 Labs	6	—	—	—
592	Qwen3 8B (Non-reasoning)	Alibaba	6	—	—	—
593	AJamba 1.5 Large	AI21 Labs	6	—	—	—
594	Nova Micro	Amazon	5.9	—	—	—
595	Granite 4.1 3B	IBM	5.9	—	—	—
596	Gemini 1.5 Flash (May '24)	Google	5.9	—	—	—
597	Claude 3 Sonnet	Anthropic	5.9	—	—	—
598	Phi-4	Microsoft	5.9	0.070	0.140	—
599	LLFM2 24B A2B	Liquid AI	5.9	—	—	—
600	Gemini 1.0 Ultra	Google	5.8	—	—	—
601	Phi-4 Multimodal Instruct	Microsoft	5.8	—	—	—
602	Mistral Large (Feb '24)	Mistral	5.8	0.500	1.50	0.050
603	Mistral Small (Sep '24)	Mistral	5.8	0.150	0.600	0.015
604	NVIDIA Nemotron Nano 12B v2 VL (Non-reasoning)	NVIDIA	5.8	—	—	—
605	Gemma 3n E4B Instruct Preview (May '25)	Google	5.8	—	—	—
606	Phi-3 Mini Instruct 3.8B	Microsoft	5.8	—	—	—
607	Qwen2.5 Coder Instruct 7B	Alibaba	5.8	—	—	—
608	OMiniCPM-V 4.6 1.3B	OpenBMB	5.7	—	—	—
609	Llama 3.2 Instruct 3B	Meta	5.7	0.050	0.330	—
610	Mixtral 8x22B Instruct	Mistral	5.7	2.00	6.00	0.200
611	Qwen1.5 Chat 110B	Alibaba	5.7	—	—	—
612	Qwen3 VL 4B Instruct	Alibaba	5.7	—	—	—
613	AJamba Reasoning 3B	AI21 Labs	5.7	—	—	—
614	Llama 2 Chat 7B	Meta	5.7	—	—	—
615	Claude 3 Haiku	Anthropic	5.6	—	—	—
616	AOlmo 3 7B Think	Allen Institute for AI	5.6	—	—	—
617	Claude 2.1	Anthropic	5.6	—	—	—
618	AMolmo 7B-D	Allen Institute for AI	5.6	—	—	—
619	RReka Flash 3	Reka AI	5.6	0.100	0.200	—
620	AOLMo 2 7B	Allen Institute for AI	5.6	—	—	—
621	GPT-3.5 Turbo	OpenAI	5.5	1.00	2.00	—
622	Mistral Medium	Mistral	5.5	—	—	—
623	DeepSeek-V2-Chat	DeepSeek	5.5	—	—	—
624	Mistral Small (Feb '24)	Mistral	5.5	0.150	0.600	0.015
625	ILing-mini-2.0	InclusionAI	5.5	—	—	—
626	DeepSeek R1 Distill Qwen 1.5B	DeepSeek	5.5	—	—	—
627	Ministral 3 8B	Mistral	5.5	—	—	—
628	Llama 3 Instruct 70B	Meta	5.5	—	—	—
629	Claude 2.0	Anthropic	5.5	—	—	—
630	LLFM 40B	Liquid AI	5.4	—	—	—
631	Qwen3.5 0.8B (Non-reasoning)	Alibaba	5.4	—	—	—
632	Qwen Chat 72B	Alibaba	5.4	—	—	—
633	PALM-2	Google	5.4	—	—	—
634	Llama 3.2 Instruct 11B (Vision)	Meta	5.4	—	—	—
635	SArctic Instruct	Snowflake	5.4	—	—	—
636	DDBRX Instruct	Databricks	5.3	—	—	—
637	SSarvam M (Reasoning, Based on Mistral Small 3.1)	Sarvam	5.3	—	—	—
638	Llama 2 Chat 70B	Meta	5.3	—	—	—
639	Llama 2 Chat 13B	Meta	5.3	—	—	—
640	CCommand-R+ (Apr '24)	Cohere	5.3	—	—	—
641	DeepSeek LLM 67B Chat (V1)	DeepSeek	5.3	—	—	—
642	Gemini 1.0 Pro	Google	5.3	—	—	—
643	DeepSeek Coder V2 Lite Instruct	DeepSeek	5.3	—	—	—
644	LExaone 4.0 1.2B (Reasoning)	LG AI Research	5.3	—	—	—
645	OOpenChat 3.5 (1210)	OpenChat	5.3	—	—	—
646	LLFM2.5-1.2B-Instruct	Liquid AI	5.2	—	—	—
647	AJamba 1.5 Mini	AI21 Labs	5.2	—	—	—
648	Qwen3 1.7B (Reasoning)	Alibaba	5.2	—	—	—
649	LLFM2.5-1.2B-Thinking	Liquid AI	5.2	—	—	—
650	LLFM2 2.6B	Liquid AI	5.2	—	—	—
651	LExaone 4.0 1.2B (Non-reasoning)	LG AI Research	5.2	—	—	—
652	AOlmo 3 7B Instruct	Allen Institute for AI	5.2	—	—	—
653	AJamba 1.7 Mini	AI21 Labs	5.2	—	—	—
654	Granite 4.0 H 1B	IBM	5.2	—	—	—
655	AJamba 1.6 Mini	AI21 Labs	5.2	—	—	—
656	Gemma 3 270M	Google	5.1	—	—	—
657	Granite 4.0 Micro	IBM	5.1	—	—	—
658	Mixtral 8x7B Instruct	Mistral	5.1	—	—	—
659	SApertus 70B Instruct	Swiss AI Initiative	5.1	—	—	—
660	NDeepHermes 3 - Llama-3.1 8B Preview (Non-reasoning)	Nous Research	5.1	—	—	—
661	Mistral 7B Instruct	Mistral	5	—	—	—
662	AMolmo2-8B	Allen Institute for AI	5	—	—	—
663	Claude Instant	Anthropic	5	—	—	—
664	Granite 4.0 1B	IBM	5	—	—	—
665	CCommand-R (Mar '24)	Cohere	5	—	—	—
666	Qwen Chat 14B	Alibaba	5	—	—	—
667	Llama 65B	Meta	5	—	—	—
668	Qwen3 1.7B (Non-reasoning)	Alibaba	4.9	—	—	—
669	Gemma 3 27B Instruct	Google	4.9	—	—	—
670	Granite 3.3 8B (Non-reasoning)	IBM	4.9	—	—	—
671	LLFM2 8B A1B	Liquid AI	4.9	—	—	—
672	Qwen3 0.6B (Non-reasoning)	Alibaba	4.8	—	—	—
673	LLFM2 1.2B	Liquid AI	4.8	—	—	—
674	Gemma 3 4B Instruct	Google	4.8	—	—	—
675	LLFM2.5-VL-1.6B	Liquid AI	4.8	—	—	—
676	Qwen3 0.6B (Reasoning)	Alibaba	4.8	—	—	—
677	Granite 4.0 350M	IBM	4.8	—	—	—
678	Ministral 3 3B	Mistral	4.8	—	—	—
679	Llama 3.2 Instruct 1B	Meta	4.8	0.027	0.201	—
680	CTiny Aya Global	Cohere	4.8	—	—	—
681	Llama 3 Instruct 8B	Meta	4.8	—	—	—
682	SApertus 8B Instruct	Swiss AI Initiative	4.8	—	—	—
683	NVIDIA Nemotron 3 Nano 4B	NVIDIA	4.8	—	—	—
684	Granite 4.0 H 350M	IBM	4.8	—	—	—
685	Gemma 3n E4B Instruct	Google	4.8	—	—	—
686	Gemma 3n E2B Instruct	Google	4.8	—	—	—
687	Gemma 3 1B Instruct	Google	4.8	—	—	—
688	Gemma 3 12B Instruct	Google	3.8	—	—	—
689	IK2 Horizon 0.9B	Institute of Foundation Models	3	—	—	—

数据来源: Artificial Analysis · 更新于 Oct 13, 2026 · PandaNpc 更新于 2026年10月13日 UTC 06:00查看完整榜单 →

指标怎么读

大模型排行榜告诉你「人更喜欢谁」，AI 基准测试告诉你「它客观上能做什么」，用量数据告诉你「开发者真的在用谁」。没有哪一张榜能单独回答「最好的 AI 模型是哪个」，所以这一页把三类都聚合进来——arena.ai 竞技场 Elo、Artificial Analysis 智能指数与语音竞技场、OpenRouter token 占比——并每日刷新。

arena.ai（原 LMArena）让用户在不知道模型身份的情况下对同一问题的两份回答二选一，再用 Elo 算法推算等级分。它衡量的是「人更喜欢哪个」，不等同于客观能力——擅长讨人喜欢的风格会占优，而严谨但啰嗦的回答可能吃亏。

票数是该模型累计参与的对战场次。新模型刚上榜时票数少、置信区间宽，排名容易波动，看排名时应连同 ± 区间一起判断。

Agent 能力榜的口径与其他榜不同：它统计的是真实 Agent 会话，而非盲投偏好。主指标「净改进」是百分比形式的提升幅度，与 Elo 不在同一量纲，不能跨榜比较。arena.ai 还给出确认成功率、可引导性、Bash 恢复等五个维度，可在完整榜单查看。

用量热度榜来自 OpenRouter，统计的是真实 API 调用的 token 份额，代表开发者用脚投票的结果，与主观偏好榜互为补充。

语音榜来自 Artificial Analysis：TTS 采用人类盲听的 Speech Arena Elo；ASR 采用 AA-WER v2，数值越低越好；原生语音对语音榜则等权综合 Big Bench Audio、Full Duplex Bench 和 τ-Voice，数值越高越好。不同榜的分数不能横向比较。来源 artificialanalysis.ai。

智能指数榜来自 Artificial Analysis：把多项公开基准测试综合成一个 0-100 的 Intelligence Index，衡量的是客观能力而非主观偏好，正好与 arena.ai 的人类盲投互补。数据来源 artificialanalysis.ai。