Picazzo Photo Booth — Turn your photos into retro AI snapshots

Picazzo — Wed, 30 Apr 2025 06:37:53 +0000

Hi devs!

I wanted to share a fun little side project I just launched: Picazzo Photo Booth — a web-based AI toy that transforms your uploaded images into vintage photo booth snapshots using OpenAI’s new gpt-image-1 model.

Why I built this

OpenAI recently released gpt-image-1 — their newest multimodal model for generating and editing images. I was curious: just how good is it at creating stylized outputs, especially ones with a nostalgic or handcrafted feel?

Picazzo was my way of testing its capabilities. Can we create charming, photobooth-style edits that feel like they came from another era — using only AI?

Originally, I wanted to convert uploaded photos into Picasso-style abstract portraits, but the results weren’t quite what I imagined. So I pivoted toward something more whimsical and accessible — retro snapshots.

I set out to build something fun, visual, and a little magical — all without requiring any manual editing or prompting from the user.

How it works

You upload a photo
The backend crafts a custom prompt to generate a 6-frame vintage snapshot strip
The image is styled and processed with high-quality settings
Boom — retro vibes in less than 5 minutes

This is a full-stack proof of concept, built with React (Tailwind CSS v4) on the frontend and Node.js + Express + SQLite on the backend. The image generation is powered by OpenAI’s gpt-image-1 API.

Sample

Built with

Cursor (yes, the AI code editor — with Gemini 2.5 Pro MAX)
OpenAI API (gpt-image-1)

I plan to open-source the project once it's a bit cleaner. For now, you can try it for free:

👉 www.picazzo.cc

Each user can generate 2 images/day (system-wide cap: 100/day)

Would love feedback!

Let me know if you hit any weird edge cases, want to see new styles added — or if you know a better prompt I should try!

✨ Thanks for reading! ✨