mirrors/stable-diffusion-webui
1
Fork 0
mirror of https://github.com/AUTOMATIC1111/stable-diffusion-webui.git synced 2026-02-01 05:11:19 -08:00
Code Issues Projects Releases Packages Wiki Activity

call torch_gc() to fix VRAM usage spike when call decode_first_stage()

Browse source
This commit is contained in:
Won-Kyu Park 2024-09-19 13:34:16 +09:00
parent 380e9a84c3
commit 71b430f703
No known key found for this signature in database
GPG key ID: 53AA79C8C9535D15
1 changed files with 1 additions and 0 deletions
Download patch file Download diff file Expand all files Collapse all files

1
modules/lowvram.py
View file

@ -53,6 +53,7 @@ def setup_for_low_vram(sd_model, use_medvram):
if module_in_gpu is not None:
module_in_gpu.to(cpu)
devices.torch_gc()
module.to(devices.device)
module_in_gpu = module