Model techniques map
Techniquesinference & servingKV cache management

specific method · filed under inference & serving

Cross-layer KV-cache reuse

Reuses KV-cache content across layers to reduce the global cache footprint.

Also called Cross-layer key-value cache reuse.

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.

cross-layer KV cache reuse in Compressed Sparse Attention 2 (CSA2) and FP4 KV caching reduce its global KV cache footprint

coreinference servingin DeepSeek-V4.1-FlashDeepSeek

Filed alongside

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