Stable Diffusion Web UI
Explore stable diffusion models using a simple browser UI.
What can I do with sd-webui?
-
The web UI allows you to import your custom models to run inference
-
Run batch-inference jobs with your models
-
Run txt2img and img2img models using the UI
-
Use your ControlNet models on the UI
-
View all your outputs and change inference parameters accordingly.
These are just a few of the most interesting features of sd-webui. For a full list, check out this url: https://github.com/AUTOMATIC1111/stable-diffusion-webui.
How does sd-webui work?
-
sd-webui requires a model for all of its features. We pre-populate a model for you.
-
Depending on the task at hand, the outputs of sd-webui are different. For instance, txt2img is selected below.
-
Txt2imge will require a text prompt and will use your model to generate an image.
-
img2img will take in a text prompt along with an image and generate a new image incorporating your text prompt.
-
There are various user-configured parameters on the UI that users can play and modify for improvements on outputs.
How to start the web UI
rubbrband web sd-webui
How to import custom models
To import a custom model that is stored at /path/to/mymodel.ckpt, copy the model into the container:
# If the model is a Stable Diffusion model
rubbrband copy-to sd-webui /path/to/mymodel.ckpt /home/engineering/stable-diffusion-webui/models/Stable-diffusion
# If the model is a ControlNet model
rubbrband copy-to sd-webui /path/to/mymodel.ckpt /home/engineering/stable-diffusion-webui/models/ControlNet
Remeber to click on the refresh button and select the copied model in the dropdown menu.