DataLearner logoDataLearnerAI
Latest AI Insights
Model Evaluations
Model Directory
Model Comparison
Resource Center
Tools
LanguageEnglish

加载中...

DataLearner logoDataLearner AI

A knowledge platform focused on LLM benchmarking, datasets, and practical instruction with continuously updated capability maps.

Products

  • Leaderboards
  • Model comparison
  • Datasets

Resources

  • Tutorials
  • Editorial
  • Tool directory

Company

  • About
  • Privacy policy
  • Data methodology
  • Contact

© 2026 DataLearner AI. DataLearner curates industry data and case studies so researchers, enterprises, and developers can rely on trustworthy intelligence.

Privacy policyTerms of service
Page navigation
目录
Model catalogQwen3.5-397B-A17B
QW

Qwen3.5-397B-A17B

Qwen3.5-397B-A17B

Release date: 2026-02-16更新于: 2026-04-02 15:49:17.9381,546
Live demoGitHubHugging FaceCompare
Parameters
397.0亿
Context length
256K
Chinese support
Supported
Reasoning ability

Data sourced primarily from official releases (GitHub, Hugging Face, papers), then benchmark leaderboards, then third-party evaluators. Learn about our data methodology

Qwen3.5-397B-A17B

Model basics

Reasoning traces
Supported
Thinking modes
Thinking Level · On (Default)Thinking Level · Off
Context length
256K tokens
Max output length
No data
Model type
多模态大模型
Release date
2026-02-16
Model file size
No data
MoE architecture
Yes
Total params / Active params
397.0B / 17B
Knowledge cutoff
No data
Qwen3.5-397B-A17B

Open source & experience

Code license
Apache 2.0
Weights license
Apache 2.0- 免费商用授权
GitHub repo
https://github.com/QwenLM/Qwen3.5
Hugging Face
https://huggingface.co/Qwen/Qwen3.5-397B-A17B
Live demo
https://chat.qwen.ai
Qwen3.5-397B-A17B

Official resources

Paper
Qwen3.5: Towards Native Multimodal Agents
DataLearnerAI blog
No blog post yet
Qwen3.5-397B-A17B

API details

API speed
3/5
💡Default unit: $/1M tokens. If vendors use other units, follow their published pricing.
Learn about pricing modes
Standard
TypeConditionInputOutput
Text-$0.500/ 1M$3.00/ 1M
Cache PricingPrompt Cache
TypeTTLWriteRead
Text5m$0.625/ 1M$0.050/ 1M
Qwen3.5-397B-A17B

Benchmark Results

Qwen3.5-397B-A17B currently shows benchmark results led by MMLU Pro (8 / 115, score 87.80), IF Bench (2 / 27, score 76.50), Pinch Bench (3 / 37, score 89.10). This page also consolidates core specs, context limits, and API pricing so you can evaluate the model from benchmark results and deployment constraints together.

Thinking
All modesThinking
Thinking mode details (1)
All thinking modesDefault (On)
Tool usage
All modesWith toolsNo tools
Internet
All modesOfflineInternet enabled

综合评估

5 evaluations
Benchmark / mode
Score
Rank/total
C-Eval
On
93
3 / 5
GPQA Diamond
On
88.40
16 / 162
MMLU Pro
On
87.80
8 / 115
HLE
On
28.70
48 / 119
HLE
OnToolsInternet
48.30
13 / 119

编程与软件工程

4 evaluations
Benchmark / mode
Score
Rank/total
LiveCodeBench
On
83.60
12 / 108
SWE-bench Verified
OnTools
76.40
16 / 93
SWE-bench Multilingual
On
69.30
5 / 7
SWE-Bench Pro - Public
On
50.90
11 / 22

多模态理解

1 evaluations
Benchmark / mode
Score
Rank/total
MMMU
On
85
4 / 28

Agent能力评测

2 evaluations
Benchmark / mode
Score
Rank/total
Pinch Bench
OnTools
89.10
3 / 37
τ²-Bench
OnTools
86.70
7 / 39

指令跟随

1 evaluations
Benchmark / mode
Score
Rank/total
IF Bench
On
76.50
2 / 27

AI Agent - 信息收集

2 evaluations
Benchmark / mode
Score
Rank/total
BrowseComp
OnToolsInternet
78.60
5 / 33
BrowseComp
OnTools
69
11 / 33

AI Agent - 工具使用

3 evaluations
Benchmark / mode
Score
Rank/total
OSWorld-Verified
OnTools
62.20
5 / 10
Terminal Bench 2.0
OnTools
52.50
13 / 28
Tool Decathlon
OnTools
38.30
3 / 4

数学推理

2 evaluations
Benchmark / mode
Score
Rank/total
AIME 2026
On
91.30
7 / 9
IMO-AnswerBench
On
80.90
7 / 7

长上下文能力

1 evaluations
Benchmark / mode
Score
Rank/total
AA-LCR
On
68.70
5 / 13
View benchmark analysisCompare with other models
Qwen3.5-397B-A17B

Publisher

阿里巴巴
阿里巴巴
View publisher details
Qwen3.5-397B-A17B

Model Overview

Qwen3.5-Plus模型就是阿里官方托管提供的Qwen3.5-397B-A17B模型。


Qwen3.5-397B-A17B模型由阿里巴巴云的Qwen团队开发,于2026年2月16日发布,作为Qwen3.5系列的首个开源权重模型。该模型作为原生视觉-语言基础模型,针对多模态代理应用的进步。

在架构和技术规格方面,它采用混合设计,将通过Gated Delta Networks的线性注意力与稀疏专家混合(MoE)结构集成,导致总参数量为3970亿,每次前向传递激活参数为170亿。上下文窗口扩展至256,000个token,便于处理推理和多模态任务中的扩展序列。预训练涉及大规模视觉-文本token,数据在中文和英文、多语言内容、STEM领域和推理元素方面丰富,并经过严格过滤。

关于核心能力和模态,该模型原生支持文本、图像和视频输入,同时生成文本输出。它在多模态推理方面表现出色,包括视觉理解、空间智能、视频分析、语言理解、代码生成以及代理工作流与工具集成,如网络搜索和代码解释器。

在性能指标上,该模型在MMLU-Pro上获得87.8分,MMLU-Redux上94.9分,SuperGPQA上70.4分,MMMU上85.0分,MMMU-Pro上79.0分,MathVision上88.6分,RealWorldQA上83.9分,VideoMME上87.5分,以及MVBench上77.6分。在比较中,它在知识、推理和编码基准上优于GLM-4.5-355B-A32B和DeepSeek-V3.2-671B-A37B等模型,同时相对于Qwen3-Max在32k和256k上下文中提供8.6x至19.0x更高的解码吞吐量,相对于Qwen3-235B-A22B提供3.5x至7.2x。

对于应用场景,它适用于自治代理系统、视觉推理、编码协助和GUI自动化。已知限制包括在超长视频处理或训练数据未覆盖的高度专业化领域中的潜在约束。

访问通过Apache 2.0许可下的开源权重分发提供,权重可在Hugging Face和GitHub等平台上获得。开发者可以通过阿里巴巴云的Bailian平台以OpenAI格式兼容的API集成它。

DataLearner 官方微信

欢迎关注 DataLearner 官方微信,获得最新 AI 技术推送

DataLearner 官方微信二维码