LCM-LoRA

LCM LoRA Colab Notebook

This repository contains the code and instructions for running the LCM LoRA models on Google Colab with a T4 GPU. These models are based on stable diffusion models with LoRA distillation, which can synthesize high-resolution images with few-step inference. You can use these models to generate realistic and diverse images from various domains, such as faces, animals, landscapes, and more.

You can Also Try on Google Colab for free & Make Sure to follow me:

Models

The LCM LoRA models include:

SDv1.5: A stable diffusion model trained on ImageNet-64 with 1.5 billion parameters.
SDXL: A stable diffusion model trained on ImageNet-128 with 3.2 billion parameters.
SSD 1B: A super-resolution diffusion model trained on ImageNet-256 with 1 billion parameters.

How to use

To use the LCM LoRA models, you need to have a Google account and access to Google Colab. You can try the models for free on Google Colab by following these steps:

1. Open the LCM LoRA Colab Notebook in your browser.

2. Click on the Connect button at the top right corner of the notebook to connect to a runtime with a T4 GPU.

4. To generate an image, type your prompt in the cell below and run it. You can also experiment with different values for the number of steps and the guidance scale.

</b>

Examples

Here are some examples of images generated by the LCM LoRA models:

Step Comparision of each Models

Support me

Hey Guys, I am a student who is passionate about AI and it’s generative models. I created this notebook to share my work and help others learn and experiment with the LCM LoRA models. If you find this notebook useful or interesting, please consider supporting me in any of the following ways:

Star this repository on GitHub and share it with your friends and colleagues.
Follow me on GitHub and Twitter for more updates and projects.
Buy me a coffee

Thank you for your support and feedback! 😊