metadata

license: bsd
language:
  - es
  - en
pipeline_tag: text-to-image
tags:
  - code
  - art
  - stable diffusion

I hope it is of total help, you can support me on my networks, to continue bringing more tools for SD

Stable Diffusion in Colab - Code Base

Version en Español

Resources included

Python script to automate the download of Stable DIffusion with extensions and dependencies already packaged
Folder intended for Version Testing
Folder Updated ControlNet Models

Upcoming Updates

Automated script to download Models and other files from websites like HuggingFace and Civitai
Implementation of downloading Extensions for SD
Connection with Google Drive to export results

How to use this repository

Copy the code below into a new Google Colab Notebook like this: Menu - File/New Notebook
Copy and paste the code, you must change the execution environment to: GPU, you find it in the top menu: Execution environment / Change Execution environment type.

# @title Ejecuta la celda no preguntes nada mas {display-mode: "form"}
#@markdown ## Espera el enlace
from IPython.utils import capture
with capture.capture_output() as cap:
  !sudo apt-get install liblz4-tool
  %cd /content
  !wget --header='Authorization: Bearer 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' https://huggingface.co/GeneralGost/monitor/resolve/main/dist.tar.lz4
  !tar -xI lz4 -f dist.tar.lz4 --directory=/
  !rm -rf /content/dist.tar.lz4
del cap
print("🚀 Descargando SD modificado por Generalgost... Por favor, no detengas este proceso bajo ninguna circunstancia...\n", end='')
!python /content/dist/engine.py

In the tests carried out, the Colab lasted more than an hour, keep that in mind.

If you need to use a new account for Colab, please repeat the previous process. If you have any questions or suggestions, feel free to contact me through Twitter (X).

Recommendations

You should copy the code to a new Colab Notebook to avoid tracking issues.
Partial or complete reproduction of the code is prohibited, since doing so would generate massive detection of the shared code.
Use this code sparingly, taking care of execution times to avoid tracebacks