Open Models Almanac
Methodsinference & servingKV cache management

single mention · inference & serving

Unified paged cache layout for KDA states and MLA KV

Stores KDA states and MLA KV in a shared paged block pool with common page sizing and allocation machinery.

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.

Filed under inference & serving › KV cache management