Release dates
— Qwen3-Next-80B-A3B released open weights. ↗
Qwen3-Next-80B-A3B
An 80B-total, 3B-active architecture preview combining hybrid linear attention and ultra-sparse MoE.
MODEL NUMBER108
- First public date
- 2025-09-11
- Organization
- Alibaba Cloud
- Family
- Qwen
01
Specifications
Only published specifications are shown.
- Total parameters
- 80B
- Active parameters
- 3B
- Architecture
- Hybrid
- Context
- 262,144 tokens
- Input modalities
- Text
- Open weights
- Yes
- License
- Apache-2.0
- Status
- Legacy
02
Overview
PreviousQwen3-235B-A22B235B · 128K ctx
CurrentQwen3-Next-80B-A3B80B · 262K ctx
04
Architecture & capabilities
Attention
Gated DeltaNetGated Attention
Model capabilities
Hybrid reasoningCodingTool useLong context
05
Release history
Qwen3-Next-80B-A3B released open weights.
View source06
Related models
Select a card to view the related model.
07
Sources
Suggest a correction
Copy this template, add the field, proposed value, and primary source, and share it with the archive maintainer.