Add activation learned w6/int8 mixed precision HW quantization of v1 test
#6
by LokkenJP - opened
This PR adds an activation learned w6/int8 mixed precision HW quantization of v1 test.
My usual recipe for a int8 convrot replacement at the same perceived quality but lower space requirements.
Actual VRAM usage depends on the workflow and other loaded models.
A full model card and standalone releases will follow only once gating for S3 is lifted.
LokkenJP changed pull request status to closed