跳转到主要内容
AITier

Qwen3.8 Max (0902)

AlibabaQwen开源权重Qwen3.8-Max License · 商用许可

描述

Qwen3.8 Max, released as the Qwen3.8-2.4T-A95B open-weight checkpoint, is Qwen's flagship mixture-of-experts model for coding, research, professional workflows, and long-horizon agents. It has 2.4 trillion total parameters with 95 billion active parameters, a native 262,144-token context window extensible to about 1 million tokens, and always-on thinking. QwenCloud serves the same model in the Qwen3.8 Max API family with text and image input.

发布日期
2026-09-02
参数规模
2.4T
上下文长度
1.0M
支持模态
image, pdf, text, video

能力雷达图

45
general
72
coding
93
reasoning
69
science
60
agents
90
multimodal

排行榜排名

领域#排名分数来源
智能体能力模型榜18
61.0
LS
代码能力榜43
92.0
AA
通用能力榜36
78.0
AA
数学推理31
91.0
LB
多模态榜1
84.0
LS
推理能力24
88.0
LB
科学能力62
80.0
AA

基准测试分数 (LLM Stats)

(LLM Stats (zeroeval))

Agents

QwenSVG1713.00 / 2000自报
AndroidWorld85.3%自报
MobileWorld77.8%自报
AndroidBench75.1%自报
CoWorkBench74.8%自报
Toolathlon72.5%自报
Workspace Bench67.7%自报
Job Bench53.4%自报
OneMillion Bench52.5%自报
Agents' Last Exam52.4%自报
MLS-Bench Lite41.0%自报
AutomationBench27.3%自报

Code

QwenReactBench1724.00 / 2000自报
PaperBench93.0%自报
QwenSWEBench80.7%自报
FrontierSWE73.5%自报
SkillsBench70.2%自报
Vision2Web69.0%自报
QwenQoderBench58.4%自报
DeepSWE 1.156.6%自报
NL2Repo55.9%自报

Healthcare

HealthBench60.2%自报

Instruction Following

IFBench82.8%自报

Knowledge

PLawBench73.2%自报
PRBench-Finance58.3%自报
PRBench-Legal57.6%自报

Long Context

MRCR v2 (8-needle)92.9%自报
LongBench v266.3%自报

Multimodal

OSWorld-Verified86.1%自报

Reasoning

GPQANYU + Cohere + Anthropic (2023)92.6%自报
Terminal-Bench 2.186.6%自报
SWE-Bench ProPrinceton NLP (2024)67.7%自报
Humanity's Last Exam (with tools, text-only)56.2%自报
Humanity's Last Exam43.6%自报

Search

WideSearch81.9%自报

Vision

VideoMME w sub.90.4%自报
RealWorldQA88.0%自报
ScreenSpot Pro84.5%自报
MMMU-Pro82.3%自报
LVBench81.8%自报
ERQA77.8%自报
PerceptionBench63.5%自报

AA 评测指数

(Artificial Analysis)
Gpqa(NYU + Cohere + Anthropic (2023))
92.8
Terminalbench V2 1
88.8
Lcr(Artificial Analysis)
80.3
Coding Index(Artificial Analysis)
76.2
Scicode(UIUC + Argonne National Lab (2024))
52.1
Tau Banking
47.8
Intelligence Index(Artificial Analysis)
45.4
Hle(Center for AI Safety + Scale AI (2025))
43.1
Terminalbench V4 0
38.9

LLM Stats 分类评分

(LLM Stats (zeroeval))
Physics
90
Biology
90
Chemistry
90
Video
90
Instruction Following
80
Multimodal
80
Search
80
Spatial Reasoning
80
Grounding
80
Vision
80
Long Context
70
Reasoning
70
Structured Output
70
General
70
Agents
70
Code
70
Productivity
60
Healthcare
60
Tool Calling
60
Math
50

定价

输入价格$2 / 1M tokens
输出价格$6 / 1M tokens
混合价格(3:1)$3 / 1M tokens
缓存读取价格$0.25 / 1M tokens
缓存写入价格$2.5 / 1M tokens

速度

Tokens/秒39.3
首Token延迟1.77s
首回答延迟52.69s

供应商价格排行

供应商价格排行

6 个供应商

最便宜: DeepInfra最贵: EmpirioLabs AI
供应商输入输出
1DeepInfra最便宜
$0
$0
2Novita
$0
$0.00001
3Fireworks
$0
$0.00001
4Together
$0
$0.00001
5Alibaba主要
$2
$6
6EmpirioLabs AI
$2
$6

比较该模型在不同 API 供应商之间的定价。

外部链接

LLM StatsArtificial Analysis