SuperMe - Generative Photo AI

Learn how to build an API that trains and generate photos featuring... you! Using FastAPI, Upstash, Replicate, Jupyter, and more

Tech Stack

Python 3.13
Django (pip install "Django>=5.1,<5.2")
Upstash - serverless redis, qstash for async endpoint scheduling, rate limiting, caching, and more.
Replicate - train and run generative ai model featuring your face
Python requests (pip install requests)
Jupyter (pip install jupyter)
Python Decouple to load environment variables (e.g. .env) with type casting and default values.
ostris/flux-dev-lora-trainer. Model made to allow you to fine-tune FLUX with your own images (pre-trained model designed for your training)

Tutorial

Download the following:

Open a command line (Terminal, VSCode Terminal, Cursor Terminal, Powershell, etc)

Clone this Repo

mkdir -p ~/dev/superme-api
cd ~/dev/superme-api
git clone https://github.com/codingforentrepreneurs/super-me-photo-ai-api .

Checkout the start branch

git checkout start

Make the code yours

rm -rf .git
git init
git add --all
git commit -m "I am the capitan now"

Create a Python vitual environment macOS/Linux/WSL

python3.12 -m venv venv
source venv/bin/activate

windows powershell

c:\Path\To\Python312\python.exe -m venv venv
.\venv\Scripts\activate

Install requirements

(venv) python -m pip install pip --upgrade
(venv) python -m pip install -r requirements.txt

To add support for .heic images (e.g. iPhone images) install libheif via homebrew:

brew install libheif
(venv) python -m pip install pillow-heif

If on linux or Docker, you can use sudo apt-get install libheif-dev