lllyasviel/stable-diffusion-webui-forge
1
0
Fork 0
mirror of https://github.com/lllyasviel/stable-diffusion-webui-forge.git synced 2026-02-23 00:03:57 +00:00
Code Issues Packages Projects Releases Wiki Activity

Update README.md

Browse Source
This commit is contained in:
lllyasviel
2024-01-25 18:15:42 -08:00
parent be26d30cbb
commit 48e0dfad6c
1 changed files with 2 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
README.md
View File

@@ -63,9 +63,9 @@ Currently, the behaviors is:
"When loading a model to GPU, Forge will decide whether to load the entire model, or to load separated parts of the model. Then, when loading another model, Forge will try best to unload the previous model."
**The only one flag that you may need** is `--disable-offload-from-vram`, to change the above behavior to
**The only one flag that you may still need** is `--disable-offload-from-vram`, to change the above behavior to
"When loading a model to GPU, Forge will decide whether to load the entire model, or to load separated parts of the model. Then, when loading another model, Forge will try best to keep the previous model in GPU without unload it."
"When loading a model to GPU, Forge will decide whether to load the entire model, or to load separated parts of the model. Then, when loading another model, Forge will try best to keep the previous model in GPU without unloading it."
You should `--disable-offload-from-vram` when and only when you have more than 20GB GPU memory, or when you are on MAC MPS.