DiffusionGPT: LLM-Driven Text-to-Image Generation System

This repository contains the pytorch codes of our paper "DiffusionGPT: LLM-Driven Text-to-Image Generation System".

Overview

Diffusion-GPT leverages Large Language Models (LLM) to offer a unified generation system capable of seamlessly accommodating various types of prompts and integrating domain-expert models.

Demo

• Spaces Demo 1
• Spaces Demo 2

Environment

pip install -r requirements.txt

Model Weights

All the checkpoints can get from link.

Get Started

python DiffusionGPT-sdxl-app.py

or

python DiffusionGPT-sd15-app.py