Update README.md
This commit is contained in:
@@ -30,3 +30,6 @@ To run this, the text enconder is what takes most of the VRAM, but can be quanti
|
||||
After that, the sampling single image at 1024 can be expected to take similar amounts than SDXL. For VAE the base SDXL VAE is used.
|
||||
|
||||

|
||||
|
||||

|
||||
|
||||
|
||||
Reference in New Issue
Block a user