Open Models Almanac
Methodsinference & servinginference kernel

implementation detail · filed under inference & serving

MoE-side chunking

Splits the token batch within the MoE kernel itself.

The sources on this page only mention it in passing. The line above is the extraction's one-line gloss. See inference kernel for the methods filed beside it.

sources
2
model
1
lab adopt it
1
strongest
used

Documented in

Evidence

2 quotes from 2 sources, one card per model, each opening with its strongest.

How sources treat it
used 2

NVIDIA-Nemotron-3-Ultra-550B-A55B

NVIDIAadopted2 sources

The fix is MoE-side chunking, splitting the token batch within the MoE kernel itself

usedsoftware implementationin vLLMNVIDIA
1 more from this model

The fix is MoE-side chunking, splitting the token batch within the MoE kernel itself

usedinference servingin Nemotron 3 UltraNVIDIA
↑