Open Models Almanac
Methodsinference & servinginference kernel

single mention · inference & serving

Single-launch fusion of attention, indexer, routing, experts, and all-reduces

Combines attention, indexer, router, MXFP4 experts, and both all-reduces in one launch.

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.

Filed under inference & serving › inference kernel