Model techniques map
Techniquesoptimizationtraining parallelism

implementation detail · filed under optimization

Redundant-Expert Capacity Reservation

An implementation detail that reserves a fixed number of redundant-expert slots per rank so planning always has a feasible solution.

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.

Reserving E/R redundant-expert slots per rank therefore guarantees that planning always admits a feasible solution, so training is never interrupted.

coresoftware implementationin Kimi K3Moonshot AI

Filed alongside

Other methods under optimization :: training parallelism.