gmongaras/Stable-Diffusion-3-From-Scratch
A repo that attempts to train stable diffusion 3 from scratch
This project provides the foundational code and scripts to train a Stable Diffusion 3 model from the very beginning, allowing researchers or hobbyists to understand and replicate the training process. You provide computational resources and raw data, and the project outputs a custom image generation model capable of creating images from text prompts. It is designed for those who want to delve into the core mechanics of large-scale image synthesis model training.
No commits in the last 6 months.
Use this if you are a machine learning researcher or engineer interested in the technical challenge of training a state-of-the-art text-to-image model from scratch, using your own infrastructure.
Not ideal if you simply want to generate images or fine-tune an existing model without undertaking the full training process yourself, as this project requires significant technical setup and computational resources.
Stars
37
Forks
4
Language
Python
License
MIT
Category
Last pushed
Jun 05, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/gmongaras/Stable-Diffusion-3-From-Scratch"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
NVlabs/Sana
SANA: Efficient High-Resolution Image Synthesis with Linear Diffusion Transformer
FoundationVision/VAR
[NeurIPS 2024 Best Paper Award][GPT beats diffusion🔥] [scaling laws in visual generation📈]...
nerdyrodent/VQGAN-CLIP
Just playing with getting VQGAN+CLIP running locally, rather than having to use colab.
huggingface/finetrainers
Scalable and memory-optimized training of diffusion models
AssemblyAI-Community/MinImagen
MinImagen: A minimal implementation of the Imagen text-to-image model