single mention · optimization
FP8 quantization of selected layers
Adding mixed-FP8 layers on top of an NVFP4-amax recipe; the evidence does not specify further algorithmic details.
The sources on this page only mention it in passing. The line above is the extraction's one-line gloss. See training precision for the methods filed beside it.
Also called Mixed-FP8 quantization, mixed-FP8.
Filed under optimization › training precision