Skip to content
This repository has been archived by the owner on Jul 12, 2024. It is now read-only.

omniinfer/sd-webui-cloud-inference

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

43 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Stable Diffusion Web UI Cloud Inference

IMAGE ALT TEXT HERE

What capabilities does this extension offer?

This extension enables faster image generation without the need for expensive GPUs and seamlessly integrates with the AUTOMAIC1111 UI.

Benefits:

  1. No expensive GPUs required, can even use the CPU.
  2. No need to change your workflow, compatible with the usage and scripts of sd-webui, such as X/Y/Z Plot, Prompt from file, etc.
  3. Support for 10000+ Checkpoint models, don't need download

Compatibility and Limitations

Feature Support Limitations
txt2img
txt2img_hires.fix
txt2img_sdxl_refiner
txt2img_controlnet
img2img
img2img_inpaint
img2img_sdxl_refiner
img2img_controlnet
extras upscale
vae model
scripts - X/Y/Z plot
scripts - Prompt matrix
scripts - Prompt from file

How it works

how it works

Guide

1. Install sd-webui-cloud-inference

image-mh (4) Screenshot 2023-07-26 at 14 40 49-mh Screenshot 2023-07-26 at 14 44 10-mh

2. Get your omniinfer.io Key

Open omniinfer.io in browser

We can choice "Google Login" or "Github Login" image

Screenshot 2023-07-26 at 14 44 10-mh

3. Enable Cloud Inference feature

Let us back to Cloud Inference tab of stable-diffusion-webui

Screenshot 2023-07-26 at 14 46 59-mh

4. Test Txt2Img

Let us back to Txt2Img tab of stable-diffusion-webui

Screenshot 2023-07-26 at 14 51 27-mh From now on, you can give it a try and enjoy your creative journey.

Furthermore, you are welcome to freely discuss your user experience, share suggestions, and provide feedback on our Discord channel.

5. Advanced - Lora

Screenshot 2023-07-26 at 14 58 33-mh (1)

7. Advanced - Img2img Inpainting

image-mh Screenshot 2023-07-25 at 21 08 28-mh image-mh (4)

8. Advanced - VAE

image

or you can use the VAE feature with X/Y/Z

Screenshot 2023-07-25 at 22 35 30-mh

9. Advanced - ControlNet

image-mh (2)Screenshot 2023-07-25 at 22 23 01-mh

9. Advanced - ControlNet

image-mh (2)Screenshot 2023-07-25 at 22 23 01-mh

10. Advanced - Upscale and Hires.Fix

Screenshot 2023-08-09 at 20 13 58-mh

Screenshot 2023-08-09 at 20 16 18-mh

11. Advanced - Model Browser

Screenshot 2023-08-27 at 17 03 48-mh

12. Advanced - Tiny Model

The AUTOMATIC1111 webui loads the model on startup. However, on low-memory computers like the MacBook Air, the performance is suboptimal. To address this, we have developed a stripped-down minimal-size model. You can utilize the following commands to enable it.

its will reduce memory from 4.8G -> 739MB

  1. Download tiny model and config to model config.
wget -O ./models/Stable-diffusion/tiny.yaml https://github.com/omniinfer/sd-webui-cloud-inference/releases/download/tiny-model/tiny.yaml
wget -O ./models/Stable-diffusion/tiny.safetensors https://github.com/omniinfer/sd-webui-cloud-inference/releases/download/tiny-model/tiny.safetensors
  1. start webui with tiny model --ckpt=/stable-diffusion-webui/models/Stable-diffusion/tiny.safetensors

13. Advanced - SDXL Refiner

Screenshot 2023-09-04 at 13 45 24-mh