Model techniques map
Techniquesinference & servingKV cache management

specific method · filed under inference & serving

Write-back external KV-cache policy

Decouples prefix retention from GPU residency using a write-back cache design.

Also called write-back design.

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.

We therefore decouple prefix retention from GPU residency with a write-back design.

coreinference servingin Kimi K3Moonshot AI

Filed alongside

Other methods under inference & serving :: KV cache management.