Model techniques map
Techniquesinference & servinginference kernel

implementation detail · filed under inference & serving

Fused AttnRes merge and RMSNorm kernel

Fuses the AttnRes output merge and partial-sum update with subsequent RMSNorm into the preceding tensor-parallel all-reduce.

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.

the merging of the AttnRes output with its partial-sum update, together with the subsequent RMSNorm, is fused into the preceding TP all-reduce

coreinference servingin Kimi K3

Filed alongside

Other methods under inference & serving :: inference kernel.