Model techniques map
Techniquesoptimizationtraining runtime

specific method · filed under optimization

Tensor-level automatic-differentiation activation checkpointing

An activation checkpointing mechanism that operates at tensor level and supports automatic differentiation.

source
1
models
2
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 implement a tensor-level activation checkpointing mechanism with automatic differentiation support.

coresoftware implementationin DeepSeek-V4DeepSeek

Filed alongside

Other methods under optimization :: training runtime.