What is the Artificial Analysis Intelligence Index?

The Artificial Analysis Intelligence Index v4.0 is a composite benchmark that aggregates performance across 10 evaluations spanning mathematics, science, coding, agentic tasks, and reasoning to measure AI capabilities holistically.

How is the Intelligence Index calculated?

The index aggregates scores from 10 benchmarks: GDPval-AA, τ²-Bench, Terminal-Bench Hard, SciCode, AA-LCR, AA-Omniscience, IFBench, Humanity's Last Exam, GPQA Diamond, and CritPt. All tests are independently run on standardized hardware.

How does this differ from LMArena?

LMArena uses crowdsourced user votes (Elo ratings) reflecting subjective preferences. The AA Intelligence Index uses standardized automated benchmarks with objective scoring across specific technical domains.

Where can I find the original data?

The original leaderboard is available at artificialanalysis.ai/leaderboards/models and the methodology at artificialanalysis.ai/evaluations/artificial-analysis-intelligence-index.

Artificial Analysis Intelligence Index AI模型智能指数排行榜

Name: Artificial Analysis Intelligence Index AI模型智能指数排行榜
Creator: DataLearner
License: https://creativecommons.org/licenses/by/4.0/

Artificial Analysis Intelligence Index v4.0 综合了10项权威评测基准（GDPval-AA、Terminal-Bench、GPQA Diamond、SciCode等），从数学、科学、编程、推理等多维度对AI模型进行全面评估和排名。

榜首模型

Claude Opus 5.5 (max with fallback)

最高得分

模型数量

263

数据版本

2026年10月01日

数据来源: Artificial Analysis

来源：全部国产模型

榜单历史快照月份:

排名总表

排名	模型名称	智能指数	机构
	Claude Opus 5.5 (max with fallback)Anthropic	58	Anthropic
	Claude Opus 5.5 (xhigh with fallback)Anthropic	56	Anthropic
	Claude Sonnet 5.5 (max with fallback)Anthropic	56	Anthropic
4	Claude Opus 5.5 (high with fallback)Anthropic	54	Anthropic
5	Claude Fable 5.1Anthropic	53	Anthropic
6	Claude Fable 5.1Anthropic	53	Anthropic
7	GPT-6 Astra (max)OpenAI	53	OpenAI
8	Gemini 4 Argon (high)Google	53	Google
9	GPT-6 Astra (xhigh)OpenAI	52	OpenAI
10	Claude Sonnet 5.5 (xhigh with fallback)Anthropic	52	Anthropic
11	GPT-6.1 Sol (max)OpenAI	52	OpenAI
12	Claude Opus 5.5 (medium with fallback)Anthropic	51	Anthropic
13	Claude Fable 5.1Anthropic	51	Anthropic
14	GPT-6.1 Sol (xhigh)OpenAI	51	OpenAI
15	GPT-6 Astra (high)OpenAI	51	OpenAI
16	GPT-6.1 Sol (high)OpenAI	50	OpenAI
17	GPT-6 Astra (medium)OpenAI	50	OpenAI
18	Claude Fable 5.1Anthropic	49	Anthropic
19	Muse Spark 1.3 (max)Meta	48	Meta
20	GPT-6.1 Sol (medium)OpenAI	48	OpenAI
21	GPT-6 Sol (max)OpenAI	48	OpenAI
22	Claude Fable 5.1Anthropic	47	Anthropic
23	Claude Sonnet 5.5 (high with fallback)Anthropic	47	Anthropic
24	Grok 4.7 (xhigh)SpaceXAI	46	SpaceXAI
25	Grok 4.7 (high)SpaceXAI	46	SpaceXAI
26	MiMo-V2.6-ProXiaomi	46	Xiaomi
27	GPT-6 Astra (low)OpenAI	46	OpenAI
28	Qwen3.8 Max (0902)Alibaba	45	Alibaba
29	Muse Spark 1.3 (xhigh)Meta	45	Meta
30	GLM-5.3 (max)智谱AI	45	智谱AI
31	GPT-6 Sol (xhigh)OpenAI	44	OpenAI
32	Step 5 PreviewStepFun	44	StepFun
33	Kimi K3 (max)Moonshot AI	44	Moonshot AI
34	GPT-6 Sol (high)OpenAI	43	OpenAI
35	Claude Opus 5.5 (low with fallback)Anthropic	42	Anthropic
36	GPT-6.1 Sol (low)OpenAI	42	OpenAI
37	GPT-5.6 Terra (max)OpenAI	42	OpenAI
38	GLM-5.3-FlashZ AI	42	Z AI
39	Gemini 3.8 Flash (high)Google	41	Google
40	Claude Sonnet 5.5 (medium with fallback)Anthropic	41	Anthropic
41	Qwen3.8 2.4T A95BAlibaba	40	Alibaba
42	Qwen3.8-Flash-NextAlibaba	40	Alibaba
43	GPT-6 Sol (medium)OpenAI	40	OpenAI
44	Gemini 3.8 Flash (medium)Google	40	Google
45	DeepSeek V4.1 Flash (max)DeepSeek	39	DeepSeek
46	GPT-5.6 Terra (xhigh)OpenAI	38	OpenAI
47	MiMo-V2.6-FlashXiaomi	38	Xiaomi
48	GPT-6 Luna (max)OpenAI	37	OpenAI
49	DeepSeek-V4-Pro (max)DeepSeek-AI	36	DeepSeek-AI
50	DeepSeek V4 Flash Vision (max)DeepSeek	35	DeepSeek
51	GLM-5.3 (low)Z AI	34	Z AI
52	GPT-5.6 Terra (high)OpenAI	34	OpenAI
53	JT-4.1 Flash 236B A21BChina Mobile	34	China Mobile
54	GPT-6 Sol (low)OpenAI	34	OpenAI
55	GPT-6 Luna (xhigh)OpenAI	34	OpenAI
56	Qwen3.8-27B (xhigh)阿里巴巴	34	阿里巴巴
57	Motif 3Motif Technologies	34	Motif Technologies
58	Gemini 3.8 Flash (low)Google	33	Google
59	GPT-5.3 Codex (xhigh)OpenAI	33	OpenAI
60	GPT-6 Luna (high)OpenAI	32	OpenAI
61	K2 Horizon 375B A23BMBZUAI Institute of Foundation Models	31	MBZUAI Institute of Foundation Models
62	GPT-5.6 Terra (medium)OpenAI	30	OpenAI
63	Kimi K3 (low)Moonshot AI	30	Moonshot AI
64	Gemini 3.1 Pro PreviewGoogle DeepMind	30	Google DeepMind
65	GPT-6 Luna (medium)OpenAI	29	OpenAI
66	MiniMax M3MiniMaxAI	29	MiniMaxAI
67	Nex-N2-ProNex AGI	28	Nex AGI
68	Solar Pro 4Upstage	28	Upstage
69	GPT-6 Sol (non-reasoning)OpenAI	28	OpenAI
70	Qwen3.8-27B (medium)阿里巴巴	28	阿里巴巴
71	GPT-5.6 Terra (low)OpenAI	27	OpenAI
72	JT-4.1 Flash 236B A21B (non-reasoning)China Mobile	27	China Mobile
73	Quasar 438B (max)Multiverse Computing	27	Multiverse Computing
74	Apodex 1.1Apodex	26	Apodex
75	Qwen3.8-27B (low)阿里巴巴	26	阿里巴巴
76	GPT-5.5 InstantOpenAI	26	OpenAI
77	Kimi K2.7 CodeMoonshot AI	26	Moonshot AI
78	Inkling SmallThinking Machines	26	Thinking Machines
79	K2 Horizon MoVA 36B A4BInstitute of Foundation Models	25	Institute of Foundation Models
80	Hy3腾讯AI实验室	25	腾讯AI实验室
81	Qwen3.7-Plus阿里巴巴	25	阿里巴巴
82	Inkling (xhigh)Thinking Machines	25	Thinking Machines
83	Solar Open2 250BUpstage	25	Upstage
84	DeepSeek V4.1 Flash (non-reasoning)DeepSeek	25	DeepSeek
85	Ling-3.0-flash-VLInclusionAI	25	InclusionAI
86	Solar Mini 4Upstage	24	Upstage
87	Nemotron 3 UltraNVIDIA	23	NVIDIA
88	Qwen2-57B-A14B阿里巴巴	23	阿里巴巴
89	Ling-3.0-flash-FinInclusionAI	23	InclusionAI
90	Gemini 3.5 Flash-LiteGoogle DeepMind	22	Google DeepMind
91	G9v3-39A5BAI9Stars	22	AI9Stars
92	KAT-Coder-Pro V2KwaiKAT	22	KwaiKAT
93	Qwen3.5 397B A17B (non-reasoning)Alibaba	21	Alibaba
94	GPT-6 Luna (low)OpenAI	21	OpenAI
95	GPT-5.6 Terra (non-reasoning)OpenAI	21	OpenAI
96	K2 Horizon 7BInstitute of Foundation Models	21	Institute of Foundation Models
97	Qwen3.5-Omni-Plus阿里巴巴	20	阿里巴巴
98	DeepSeek V4 Pro 0813 (non-reasoning)DeepSeek	20	DeepSeek
99	OpenAI o3OpenAI	20	OpenAI
100	Qwen3.8 27B (non-reasoning)Alibaba	20	Alibaba
101	Ling 3.0 FlashInclusionAI	20	InclusionAI
102	K-EXAONE 2.0LG AI Research	20	LG AI Research
103	LongCat 2.0LongCat	19	LongCat
104	JT-35B-FlashChina Mobile	19	China Mobile
105	Qwen3.5-397B-A17B阿里巴巴	18	阿里巴巴
106	GPT-6 Luna (non-reasoning)OpenAI	18	OpenAI
107	Qwen3.6-35B-A3B阿里巴巴	18	阿里巴巴
108	Qwen3.5 122B A10B (non-reasoning)Alibaba	18	Alibaba
109	Muse Glimmer-30B (high)Facebook AI研究实验室	17	Facebook AI研究实验室
110	Doubao Seed CodeByteDance Seed	17	ByteDance Seed
111	Haiku 4.5Anthropic	17	Anthropic
112	Gemma 4 26B A4BGoogle DeepMind	17	Google DeepMind
113	Ring-2.6-1TInclusionAI	17	InclusionAI
114	K2 Horizon 3.7BInstitute of Foundation Models	16	Institute of Foundation Models
115	Qwen3.5-122B-A10B阿里巴巴	16	阿里巴巴
116	Haiku 4.5 (non-reasoning)Anthropic	15	Anthropic
117	Qwen3.6 35B A3B (non-reasoning)Alibaba	15	Alibaba
118	Granite 4.2 30BIBM	15	IBM
119	Gemma 4 31BGoogle DeepMind	15	Google DeepMind
120	ERNIE 5.0 Thinking Preview百度	14	百度
121	Mistral Medium 3.5MistralAI	14	MistralAI
122	Gemma 4 12BGoogle	14	Google
123	Nova 2 Pro（Preview） (medium)亚马逊	14	亚马逊
124	Gemma 4 31B (non-reasoning)Google	14	Google
125	Nova 2 Omni（Preview） (medium)亚马逊	14	亚马逊
126	Apriel-v1.6-15B-ThinkerServiceNow	13	ServiceNow
127	Nova 2 Lite (high)亚马逊	13	亚马逊
128	Qwen3.5 9B (non-reasoning)Alibaba	13	Alibaba
129	EXAONE 4.5 33BLG AI Research	13	LG AI Research
130	Command A+Cohere	13	Cohere
131	Gemma 4 26B A4B (non-reasoning)Google	13	Google
132	Qwen3.5 4BAlibaba	13	Alibaba
133	Nemotron 3.5 LightningNVIDIA	13	NVIDIA
134	Nemotron 3 SuperNVIDIA	13	NVIDIA
135	Nova 2 Pro（Preview） (low)亚马逊	13	亚马逊
136	Nova 2 Lite (medium)亚马逊	12	亚马逊
137	Qwen3.5-Omni-Flash阿里巴巴	12	阿里巴巴
138	MiniCPM5-2BOpenBMB	12	OpenBMB
139	Mercury 2.5Inception	12	Inception
140	JT-MINIChina Mobile	12	China Mobile
141	Magistral Medium 1.2Mistral	12	Mistral
142	Nova 2 Lite (low)亚马逊	12	亚马逊
143	HyperNova 60B 2605 (high)Multiverse Computing	12	Multiverse Computing
144	Nemotron Cascade 2 30B A3BNVIDIA	12	NVIDIA
145	GPT OSS 120B (high)OpenAI	12	OpenAI
146	K2 Think V2MBZUAI	11	MBZUAI
147	LongCat Flash LiteLongCat	11	LongCat
148	HyperCLOVA X SEED Think (32B)Naver	11	Naver
149	Mistral Small 4Mistral	11	Mistral
150	Qwen3-Next阿里巴巴	11	阿里巴巴
151	Qwen3.5-9B阿里巴巴	11	阿里巴巴
152	Nova 2 Omni（Preview） (low)亚马逊	11	亚马逊
153	Granite 4.2 8BIBM	11	IBM
154	Ling 3.0 TinyInclusionAI	11	InclusionAI
155	Mi:dm K 2.5 ProKorea Telecom	11	Korea Telecom
156	G9v3-3BAI9Stars	11	AI9Stars
157	Trinity Large ThinkingArcee AI	11	Arcee AI
158	Qwen3.5 4B (non-reasoning)Alibaba	11	Alibaba
159	INTELLECT-3Prime Intellect	11	Prime Intellect
160	Solar Open 100BUpstage	10	Upstage
161	Qwen3-Omni-30B-A3B阿里巴巴	10	阿里巴巴
162	GPT OSS 120B (low)OpenAI	10	OpenAI
163	Llama 4 MaverickFacebook AI研究实验室	10	Facebook AI研究实验室
164	Nova 2.0 Pro Preview (non-reasoning)Amazon	10	Amazon
165	GPT OSS 20B (low)OpenAI	10	OpenAI
166	North Mini CodeCohere	10	Cohere
167	K2-V2 (high)MBZUAI	10	MBZUAI
168	Qwen3-Next阿里巴巴	10	阿里巴巴
169	DiffusionGemma 26B A4BGoogle	10	Google
170	Gemma 4 12B (non-reasoning)Google	9	Google
171	Mistral Large 3MistralAI	9	MistralAI
172	Qwen3-Coder-Next阿里巴巴	9	阿里巴巴
173	Motif-2-12.7BMotif Technologies	9	Motif Technologies
174	Nova PremierAmazon	9	Amazon
175	Granite 4.2 3BIBM	9	IBM
176	K2-V2 (medium)MBZUAI	9	MBZUAI
177	Mistral Small 4 (non-reasoning)Mistral	9	Mistral
178	Tri-21B-ThinkTrillion Labs	9	Trillion Labs
179	GPT OSS 20B (high)OpenAI	9	OpenAI
180	Gemma 4 E4BGoogle DeepMind	9	Google DeepMind
181	Nemotron 3 NanoNVIDIA	9	NVIDIA
182	MiniCPM5-1BOpenBMB	9	OpenBMB
183	Sarvam 105B (high)Sarvam	9	Sarvam
184	Nova 2.0 Lite (non-reasoning)Amazon	9	Amazon
185	MiniCPM5-1B (non-reasoning)OpenBMB	9	OpenBMB
186	Magistral Small 1.2Mistral	9	Mistral
187	Nanbeige4.1-3BNanbeige	8	Nanbeige
188	LFM2.5-2.6BLiquid AI	8	Liquid AI
189	EXAONE 4.0 32BLG AI Research	8	LG AI Research
190	Nova 2.0 Omni (non-reasoning)Amazon	8	Amazon
191	Llama 4 ScoutFacebook AI研究实验室	8	Facebook AI研究实验室
192	Hermes 4 70BNous Research	8	Nous Research
193	Falcon-H1R-7BTII UAE	8	TII UAE
194	Gemma 4 E2BGoogle DeepMind	8	Google DeepMind
195	Qwen3-Omni-30B-A3B阿里巴巴	8	阿里巴巴
196	Step3 VL 10BStepFun	8	StepFun
197	Llama3.3-70B-InstructFacebook AI研究实验室	8	Facebook AI研究实验室
198	Llama Nemotron UltraNVIDIA	8	NVIDIA
199	ERNIE-4.5-300B-A47B百度	8	百度
200	Hermes 4 405BNous Research	7	Nous Research
201	NVIDIA Nemotron Nano 12B v2 VLNVIDIA	7	NVIDIA
202	Gemma 4 E4B (non-reasoning)Google	7	Google
203	NVIDIA Nemotron Nano 9B V2NVIDIA	7	NVIDIA
204	Hermes 4 405B (non-reasoning)Nous Research	7	Nous Research
205	Nemotron 3 Nano 4BNVIDIA	7	NVIDIA
206	K2-V2 (low)MBZUAI	7	MBZUAI
207	Kimi Linear 48B A3B InstructKimi	7	Kimi
208	Llama3.1-405BFacebook AI研究实验室	7	Facebook AI研究实验室
209	LFM2.5-8B-A1BLiquid AI	7	Liquid AI
210	Ring-flash-2.0InclusionAI	7	InclusionAI
211	Olmo 3.1 32B ThinkAI2	7	AI2
212	C4AI Command A (202503)CohereAI	7	CohereAI
213	Qwen3.5 2BAlibaba	7	Alibaba
214	Nemotron 3 Nano (non-reasoning)NVIDIA	7	NVIDIA
215	NVIDIA Nemotron Nano 9B V2 (non-reasoning)NVIDIA	7	NVIDIA
216	Hermes 4 70B (non-reasoning)Nous Research	7	Nous Research
217	Sarvam 30B (high)Sarvam	7	Sarvam
218	Olmo 3.1 32B InstructAI2	6	AI2
219	Gemma 4 E2B (non-reasoning)Google	6	Google
220	R1 1776Perplexity	6	Perplexity
221	Llama 3.2-Vision-90BFacebook AI研究实验室	6	Facebook AI研究实验室
222	Celeris-1Celeris	6	Celeris
223	Phi-4-mini-instruct (3.8B)Microsoft Azure	6	Microsoft Azure
224	EXAONE 4.0 32B (non-reasoning)LG AI Research	6	LG AI Research
225	Qwen3.5 2B (non-reasoning)Alibaba	6	Alibaba
226	Qwen3.5 0.8BAlibaba	6	Alibaba
227	DeepHermes 3 - Mistral 24B (non-reasoning)Nous Research	6	Nous Research
228	Jamba 1.7 LargeAI21 Labs	6	AI21 Labs
229	Granite 4.0 H SmallIBM	6	IBM
230	Ministral 3 14BMistralAI	6	MistralAI
231	Qwen3-Omni-30B-A3B阿里巴巴	6	阿里巴巴
232	LFM2 24B A2BLiquid AI	6	Liquid AI
233	Phi-4-reasoningMicrosoft Azure	6	Microsoft Azure
234	Amazon Nova Micro亚马逊	6	亚马逊
235	NVIDIA Nemotron Nano 12B v2 VL (non-reasoning)NVIDIA	6	NVIDIA
236	Phi-4-multimodal-instruct Microsoft Azure	6	Microsoft Azure
237	MiniCPM-V 4.6 1.3BOpenBMB	6	OpenBMB
238	Jamba Reasoning 3BAI21 Labs	6	AI21 Labs
239	Reka Flash 3Reka AI	6	Reka AI
240	Olmo 3 7B ThinkAI2	6	AI2
241	Molmo 7B-DAllen Institute for AI	6	Allen Institute for AI
242	Ministral 3 8BMistralAI	5	MistralAI
243	Llama 3.2-Vision-11BFacebook AI研究实验室	5	Facebook AI研究实验室
244	Qwen3.5 0.8B (non-reasoning)Alibaba	5	Alibaba
245	Exaone 4.0 1.2BLG AI Research	5	LG AI Research
246	Olmo 3 7BAI2	5	AI2
247	Exaone 4.0 1.2B (non-reasoning)LG AI Research	5	LG AI Research
248	LFM2.5-1.2B-ThinkingLiquid AI	5	Liquid AI
249	Jamba 1.7 MiniAI21 Labs	5	AI21 Labs
250	LFM2.5-1.2B-InstructLiquid AI	5	Liquid AI
251	Granite 4.0 H 1BIBM	5	IBM
252	Gemma 3-270MGoogle DeepMind	5	Google DeepMind
253	Apertus 70B InstructSwiss AI	5	Swiss AI
254	Granite 4.0 MicroIBM	5	IBM
255	DeepHermes 3 - Llama-3.1 8B (non-reasoning)Nous Research	5	Nous Research
256	Molmo2-8BAI2	5	AI2
257	Ministral 3 3BMistral	5	Mistral
258	LFM2.5-VL-1.6BLiquid AI	5	Liquid AI
259	Granite 4.0 350MIBM	5	IBM
260	Tiny Aya GlobalCohere	5	Cohere
261	Apertus 8B InstructSwiss AI	5	Swiss AI
262	Granite 4.0 H 350MIBM	5	IBM
263	K2 Horizon 0.9BInstitute of Foundation Models	3	Institute of Foundation Models

数据仅供参考，以官方来源为准。模型名称旁的链接可跳转到 DataLearner 模型详情页。

评测基准组成（Intelligence Index v4.0）

Intelligence Index 综合10项严格的评测基准，全面衡量AI模型能力，避免单一维度的过拟合。

GDPval-AA

智能体真实任务

τ²-Bench

智能体工具调用

Terminal-Bench

智能体编程

SciCode

编程能力

AA-LCR

长上下文推理

AA-Omniscience

知识与幻觉检测

IFBench

指令遵循

Humanity's Last Exam

推理与知识

GPQA Diamond

科学推理

CritPt

物理推理

常见问题 (FAQ)

什么是 Artificial Analysis Intelligence Index？▼

Artificial Analysis Intelligence Index v4.0 是一个综合评测指数，聚合了10项具有挑战性的评估——涵盖数学、科学、编程、智能体任务和推理——以全面衡量AI能力。它旨在防止单一维度的过拟合，提供一个统一分数来追踪模型进步。

智能指数是如何计算的？▼

该指数综合了10项评测的分数：GDPval-AA（智能体真实任务）、τ²-Bench（工具调用）、Terminal-Bench Hard（智能体编程）、SciCode（编程）、AA-LCR（长上下文推理）、AA-Omniscience（知识与幻觉检测）、IFBench（指令遵循）、Humanity's Last Exam（推理）、GPQA Diamond（科学推理）和 CritPt（物理推理）。所有测试由 Artificial Analysis 在标准化硬件上独立运行。

这与 LMArena 排行榜有什么区别？▼

LMArena 排名基于众包用户投票（盲测A/B对比的Elo评分），反映主观的人类偏好。而 Artificial Analysis Intelligence Index 使用标准化的自动评测基准进行客观评分，衡量特定领域的技术能力。两者各有价值——LMArena 捕捉真实用户体验，而 AA Intelligence Index 提供可复现的技术测量。

在哪里可以找到原始数据？▼

原始排行榜和详细方法论可在 artificialanalysis.ai 查看。Intelligence Index 的方法论详见 Intelligence Index 页面。