From the course: Learning ComfyUI for Stable Diffusion

Unlock this course with a free trial

Join today to access over 24,900 courses taught by industry experts.

Inpainting with a generic diffusion model

Inpainting with a generic diffusion model - Stable Diffusion Tutorial

From the course: Learning ComfyUI for Stable Diffusion

Start my 1-month free trial Buy for my team

Inpainting with a generic diffusion model

“

- [Instructor] So far, we've used a specialized inpainting model. Now let's look at inpainting using any generic diffusion model. I've got a background image here that I created using Flux.1 Schnell. It's in the Exercise Files input. I want to remove this table. So this is a good case for inpainting. However, I can't use Flux.1 Schnell for inpainting because it's a distilled model, and it actually just does not work with inpainting at all. I'll have to use a different model. In an ideal world, I would use Flux.1 Dev, but that's for non-commercial applications only, so I can't use it for this course. Instead, I'm going to use Juggernaut XL, and that's going to change the quality of the image a little bit, but actually I think it will improve it slightly. Let's take a look. Here's my workflow, and I've built it out mostly. I've got that image loaded into a Load Image node, and I've already painted out the area that I want to remove, once again, by right clicking and choosing Open in…

Contents