Model techniques map
Techniquesmodel architecturetoken mixersparse attention

implementation detail · filed under model architecture

QSA micro-block compression at ratio 4

A Qwen Sparse Attention variant that compresses micro-blocks at a ratio of four.

source
1
model
1
lab adopt it
1
strongest
core

How sources treat it

One count per evidence span, weakest treatment to strongest.

core 1

Documented in

Evidence

1 span quoted from the sources, strongest treatment first.

uses compressed QSA with a compression ratio of 4, or c4.

coreunclearin Qwen3.8-Flash-NextQwen

Filed alongside

Other methods under model architecture :: token mixer :: sparse attention.