InvokeAI is a leading creative engine for Stable Diffusion models, empowering professionals, artists, and enthusiasts to generate and create visual media using the latest AI-driven technologies. The solution offers an industry leading WebUI, beautiful UI/UX, a unified canvas similar to programs such as Photoshop, and has tremendous capabilities for both inpainting and outpainting. Safe to say, we rather like this platform.
Invoke's documentation is here: https://invoke-ai.github.io/InvokeAI/features/
Youtube: https://www.youtube.com/@invokeai/videos
Prompting in Invoke
Note that Invoke has a different prompt engine from Automatic1111, which means what works well in Automatic may not work well in Invoke. You may find you enjoy the differences. Sometimes it is best to start with an image generated in Invoke rather than generating an image in Automatic1111 and bringing it over.
Workflow
An example of a workflow in Invoke:
- Generate a 512x512 image with a custom model
- Inpaint to touch up any oddities or add in key elements
- Outpaint the image to expand the size. This is important to do before any upscaling.
- Inpaint new details, clean up any seams
- When you feel the image is good in a rough state, upscale the image
- Use the additional size of the image to add in even more small details with inpainting.
This is a great way to work and create a highly detailed image, crafted to perfection.
