Training a Style LoRA in ComfyUI

Video Lecture

Section	Video Links
Training a Style LoRA in ComfyUI

Description

We will create a style LoRA. We can create any style that we can imagine.

I will demonstrate creating a LoRA that can produce crochet styled objects.

I will also use Runpod so that I am not using up all my local computer resources for a couple of hours while the LoRA is built.

Success

Register at Runpod. New users earn a small bonus credit ($5 typical, up to $500 possible) after spending $10.

DataSet

Download and extract these images into your ComfyUI/input folder.

SBCODE-Crochet-Style.zip

Models

• Download sd_xl_base_1.0_0.9vae.safetensors from https://huggingface.co/stabilityai/stable-diffusion-xl-base-1.0/tree/main
• Download realismEngineSDXL_v30VAE.safetensors from https://huggingface.co/poor7/Models/tree/main

📂 ComfyUI/
├── 📂 models/
│   ├── 📂 checkpoints/
│   │   ├── sd_xl_base_1.0_0.9vae.safetensors
│   │   └── realismEngineSDXL_v30VAE.safetensors

WGET Commands

If you are using Runpod, or a similar hosted GPU service, then you can access your running pod/instance using a terminal.

# CD into ./ComfyUI/models/checkpoints/ folder
wget -c https://huggingface.co/stabilityai/stable-diffusion-xl-base-1.0/resolve/main/sd_xl_base_1.0_0.9vae.safetensors
wget -c https://huggingface.co/poor7/Models/resolve/main/realismEngineSDXL_v30VAE.safetensors

Wait for files to download fully before running your workflows.