Vladimir Mandic 95bb8b6cdc update
2023-01-22 12:16:32 -05:00
2023-01-22 12:16:32 -05:00
2023-01-05 16:58:00 -05:00
2023-01-08 13:20:44 -05:00
2023-01-06 12:15:25 -05:00
2023-01-22 12:16:32 -05:00
2023-01-06 11:37:56 -05:00
2023-01-08 13:20:44 -05:00
2023-01-22 10:24:04 -05:00
2023-01-06 12:15:25 -05:00
2023-01-06 12:15:25 -05:00
2023-01-19 01:26:22 -05:00
2023-01-22 10:23:36 -05:00
2023-01-21 10:40:37 -05:00
2023-01-15 12:00:37 -05:00
2023-01-19 17:55:31 -05:00
2023-01-22 12:16:32 -05:00
2023-01-18 20:18:08 -05:00
2023-01-15 09:24:48 +03:00
2023-01-22 10:23:36 -05:00
2023-01-18 08:23:48 -05:00
2023-01-21 08:36:07 +03:00
2023-01-21 10:40:37 -05:00
2023-01-22 09:02:08 -05:00
2023-01-22 10:31:46 -05:00
2023-01-12 11:45:09 -05:00
2023-01-20 08:26:59 -05:00
2023-01-20 15:31:27 +01:00

Stable Diffusion - Automatic

Heavily opinionated custom fork of https://github.com/AUTOMATIC1111/stable-diffusion-webui


Notes

Fork is as close as up-to-date with origin as time allows
All code changes are merged upstream whenever possible

Fork adds extra functionality:

  • Ships with additional extensions
    e.g. System Info, Steps Animation, etc.
  • Ships with set of CLI tools that rely on SD API for execution:
    e.g. generate, train, bench, etc.
    Full list

Fork does differ in few things:

  • Additional simplified start script

    ./automatic.sh
    ./secure.sh

    Use secure.sh if you want to startup with username/password auth enabled and listen on global IP
    Existing webui.sh still exists for fresh installs to auto-install dependencies, etc.

  • Additional simplified install script

    ./install.sh

  • Drops compatibility with python 3.7 and requires 3.10

  • Updated Python libraries to latest known compatible versions
    e.g. accelerate, transformers, numpy, etc.

  • Includes opinionated System and Options configuration
    e.g. samplers, upscalers, etc.

  • Includes reskinned UI
    Black and orange dark theme with fixed width options panels and larger previews

  • Includes SD2 configuration files

  • Uses simplified folder structure
    e.g. /train, /outputs/*

  • Modified training templates

Only Python library which is not auto-updated is PyTorch itself as that is very system specific
For some Torch optimizations notes, see Wiki

Fork is compatible with regular PyTorch 1.13 as well as pre-release of PyTorch 2.0
See Wiki for Torch optimization notes


Docs

Everything is in Wiki
Except my current TODO

S
Description
Opinionated fork/implementation of Stable Diffusion
Readme 110 MiB
Languages
Python 94%
CSS 3.1%
TypeScript 2.4%
JavaScript 0.4%