Stumble onto art you've never seen. Spin up a random artist from 45,000+ names across the world's great modern collections, instantly see their work, read a one-line bio, and dive deeper — then take that spark wherever your imagination goes.
No sign-up. No install. Press Space and start exploring.
I made this for myself, as a creative engine for AI art.
Text-to-image models (Midjourney, DALL·E, Stable Diffusion, Gemini…) are incredible at channeling a specific artist's style — but you can only prompt the names you already know, and everyone reaches for the same ten: Van Gogh, Picasso, Monet, repeat. The result is a million identical-looking AI images.
So I built a machine for discovering artists you've never heard of — 45,000 of them — and the moment one of their galleries grabs me, I drop their name straight into a prompt. Suddenly you're making art in the voice of a forgotten Czech Art Nouveau illustrator or a Meiji-era woodblock master instead of the same recycled five. New names → genuinely new-looking art.
- Discover a random artist (or filter by era / nationality).
- Look — the gallery loads their actual work; the bio tells you who they were.
- Steal the vibe — drop their name into your favourite image model:
"a quiet harbor at dawn, in the style of Kuwagata Keisai, ukiyo-e woodblock, soft dawn palette"
- "More like this" to stay in that lane, or spin again for a totally fresh direction.
It turns "I don't know what to make today" into an endless well of starting points.
- 🎲 45,000+ artists — merged from the MoMA and The Met open collections, deduped and cleaned.
- 🖼️ Instant artwork gallery — see each artist's real work the moment you land on them.
- 📖 One-tap context — a Wikipedia mini-bio + a "Nationality · years" line, so every name means something.
- 🎚️ Smart filters — narrow by nationality, era, or gender, with a live count of the pool.
- ✨ "More like this" — reroll within the current artist's nationality or era to chase a style you're feeling.
- 🔗 Shareable links — every artist gets its own URL (
?artist=…) + a Share button. Send a friend a painter. - 🎨 Artist of the Day — one featured pick that's the same for everyone, every day.
- 🕑 History that sticks — your recent discoveries persist between visits.
- 🌙 Dark mode — auto-matches your system, with a manual toggle.
- ⌨️ Keyboard-fast —
Spacefor the next artist,Backspaceto go back. - 📱 Built for mobile — collapsible filters, a thumb-friendly bottom bar, zoom-proof controls.
Pure front-end, zero dependencies, zero backend — it just runs.
- Vanilla HTML / CSS / JS in a single
index.html(no framework, no build step). artists.json— a lean, pre-processed dataset (~45k artists with nationality, gender, and birth/death years) merged from the MoMA and Met Open Access collections.- Wikipedia REST API for live bios (no key, CORS-friendly).
- Hosted free on GitHub Pages.
random-artist-exploration/
├── index.html # the whole app — HTML + CSS + JS
├── artists.json # 45k artists (name, nationality, gender, born, died)
└── README.md
git clone https://github.com/Ericdataplus/random-artist-exploration.git
cd random-artist-exploration
# serve it (any static server works — needs fetch() of artists.json)
python -m http.server 8000
# then open http://localhost:8000Artist data from the open collections of The Museum of Modern Art (MoMA) and The Metropolitan Museum of Art. Bios from Wikipedia. Built with curiosity — go find an artist nobody's prompting yet. 🎨