← Back to Portfolio

tiny-stable-diffusion

2025.12 - 2026.02

Project Overview

Implemented a from-scratch Stable Diffusion 3 style pipeline with transformer-based MMDiT to generate 64x64 text-conditioned images.

Problem Definition & Goals

Key Features & Contributions

Technical Challenges & Solutions

Results & Learnings

Technologies

Python PyTorch Hugging Face Datasets Stable Diffusion 3 MMDiT Rectified Flow VAE

Links