Open Models Almanac
Methodsinference & servinginference quantization

single mention · inference & serving

Gradient-based per-layer mixed-precision quantization search

A mixed-precision method that uses gradient-based search to choose precision per layer.

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 AutoQuant search.

Filed under inference & serving › inference quantization