Open Models Almanac
Methodsinference & servinginference quantization

single mention · inference & serving

Heuristic mixed per-layer precision quantization

A heuristic per-layer mixed-precision recipe informed by sensitivity analysis.

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

Also called heuristic mixed per-layer precision recipe.

Filed under inference & serving › inference quantization