Open Models Almanac
Methodsinference & servingKV cache management

single mention · inference & serving

Overlapped host-memory prefetching

Overlaps host-memory prefetching with first-layer computation.

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

Also called host-memory prefetching.

Filed under inference & serving › KV cache management