Excellent Work - Made auto installers for Windows (run locally), RunPod and Massed Compute (run on cloud) - installs into venv , Python 3.10

#5
by MonsterMMORPG - opened

Excellent work you did thank you so much

This was requested by me so i made auto installers for Windows, RunPod and Massed Compute

Installer files : https://www.patreon.com/posts/115233922

On Windows you need to have Python 3.10.11, CUDA, Git, and C++ tools installed : step by step tutorial for these > https://youtu.be/DrhUHnYfwC0

On Massed Compute we use official SECourses image and on RunPod we use official PyTorch 2.2 template

VRAM usages

When offload_model is enabled (checked) on the Gradio interface, it uses 5.4 GB VRAM, 2x slower

When offload_model is not used (not checked) it uses 12.2 GB VRAM

When separate_cfg_infer is not checked, and offload_model is not checked, it uses 18.7 GB VRAM

The Gradio APP currently same as yours here but once your project matured I plan to add missing a lot of features hopefully like auto save generated images, generate more number of images, maybe auto try some configurations

18

17

02

16

Sign up or log in to comment