Generative Art Studio 🎨
Every image here is grown, not drawn: thousands of particles drift through an invisible field of mathematical wind, leaving trails as they go. Change the zoom, the swirl, or the palette and a completely different artwork emerges. When you love one, save it — it never existed before and will never exist again.
Thousands of particles following an invisible mathematical wind. Tweak, regenerate, and save anything you like — it is yours.
Generative art: the friendly face of algorithms
Long before diffusion models, artists like Vera Molnár were making art with algorithms — writing simple rules and letting computation surprise them. The flow field you are playing with is that same tradition: a noise function assigns every point on the canvas an angle, particles obey it, and complexity blooms out of a rule you could write on an index card.
It is also a gentle on-ramp to computational thinking. Zoom controls the noise frequency, swirl multiplies the angle range, and the particle count trades detail for chaos. You are doing parameter tuning — the daily work of machine learning engineers — except your loss function is “ooh, pretty.”
Frequently asked questions
Is this AI-generated art?
It's algorithmic (generative) art — a deterministic system of rules plus randomness, the artistic tradition that preceded and inspired today's AI image models. Everything is computed live in your browser.
Can I use the images I create?
Yes — anything you generate and save is yours to use however you like, personal or commercial.
Why does it look different every time?
Each 'New field' rolls a fresh random seed for the noise function, which reshapes the invisible wind the particles follow. Same rules, different universe.
Enjoyed playing? Learn how it works.
From flow fields to diffusion models — learn how machines went from following rules to painting portraits.
Disclosure: AI Fun Lab is a Coursera affiliate. If you enroll through these links we may earn a commission at no extra cost to you. It keeps the games free.