DataLearner logoDataLearnerAI
Latest AI Insights
Model Evaluations
Model Directory
Model Comparison
Resource Center
Tool Directory

加载中...

DataLearner logoDataLearner AI

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

产品

  • Leaderboards
  • 模型对比
  • Datasets

资源

  • Tutorials
  • Editorial
  • Tool directory

关于

  • 关于我们
  • 隐私政策
  • 数据收集方法
  • 联系我们

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

隐私政策服务条款
Page navigation
目录
Model catalogQwen3-Next
QW

Qwen3-Next

Qwen3-Next-80B-A3B

Release date: 2025-09-11更新于: 2025-09-14 21:48:381,840
Live demoGitHubHugging FaceCompare
Parameters
800.0亿
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-Next

Model basics

Reasoning traces
Not supported
Context length
256K tokens
Max output length
32768 tokens
Model type
聊天大模型
Release date
2025-09-11
Model file size
No data
MoE architecture
Yes
Total params / Active params
800.0B / 30B
Knowledge cutoff
No data
Inference modes
常规模式（Non-Thinking Mode）
Qwen3-Next

Open source & experience

Code license
Apache 2.0
Weights license
Apache 2.0- 免费商用授权
GitHub repo
GitHub link unavailable
Hugging Face
https://huggingface.co/Qwen/Qwen3-Next-80B-A3B-Instruct
Live demo
https://huggingface.co/spaces/Qwen/Qwen3-Demo
Qwen3-Next

Official resources

Paper
Qwen3-Next: Towards Ultimate Training & Inference Efficiency
DataLearnerAI blog
模型不能停，阿里又又又又要开源新模型：Qwen3-Next-80B-A3B
Qwen3-Next

API details

API speed
4/5
💡Default unit: $/1M tokens. If vendors use other units, follow their published pricing.
Standard pricingStandard
ModalityInputOutput
Text$0.5$6
Qwen3-Next

Benchmark Results

Thinking

综合评估

5 evaluations
Benchmark / mode
Score
Rank/total
BBHNormal
87.13
9 / 18
MMLUNormal
84.72
37 / 59
MMLU ProNormal
66.05
86 / 112
LiveBenchThinking
64.57
34 / 52
GPQANormal
43.43
9 / 13

数学推理

3 evaluations
Benchmark / mode
Score
Rank/total
GSM8KNormal
90.30
12 / 24
AIME2025Normal
69.50
74 / 105
MATHNormal
62.36
28 / 41

编程与软件工程

1 evaluations
Benchmark / mode
Score
Rank/total
LiveCodeBenchNormal
56.60
61 / 103

写作和创作

1 evaluations
Benchmark / mode
Score
Rank/total
Creative WritingNormal
85.30
9 / 22

Agent能力评测

1 evaluations
Benchmark / mode
Score
Rank/total
Aider-PolyglotNormal
49.80
24 / 26
查看评测深度分析与其他模型对比
Qwen3-Next

Publisher

阿里巴巴
阿里巴巴
View publisher details
Qwen3-Next-80B-A3B

Model Overview

Qwen3-Next是9月9日阿里在推特上泄露的Qwen3家族的新模型，根据爆料显示，该模型总参数800亿，但是激活参数只有30亿，是一个极其稀疏的MoE架构模型。它的效果比Qwen3-32B更好，但是吞吐量是Qwen3-32B的10倍，推理速度也快很多，是阿里Qwen系列的全新架构。


关于Qwen3-Next-80B-A3B模型更多的信息和介绍，参考DataLearnerAI的博客介绍： https://www.datalearner.com/blog/1051757449442911 

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