Playground
CommunityJev Demos & Experiments
1 Million Emojis — Chris Wijnia
A shared 1,000 by 1,000 emoji canvas where visitors paint and Jev paints alongside them: after each stroke it reads what is around it by name, weighs emoji that belong with it, and paints one or finishes a shape it is sure you meant.
- Format
- Playground
- Reading time
- 2 min
- Published by
- Community
- Author
- Chris Wijnia
- Last verified
- Oct 2026
Works forJev for Classification
Used inJev for Software & DevTools
- Platforms
- Browser
- Capabilities
- Open source
Highlights
- A thousand by a thousand cells, open to everyone; other visitors' strokes arrive live as they land.
- Each visitor gets 300 cells of ink, with five coming back every second; pick an emoji and drag to paint.
- Jev joins after each stroke: it reads what is around it by name and weighs emoji that belong with it before painting one.
- When it is sure a shape was meant, Jev finishes it; the canvas's rules and Jev's painter are open source in cwdx/1-million-emojis (MIT).
Watch out
Grounded on the live page text captured on 2026-10-15; the capture is mostly interface text (coordinates, ink counters, activity labels), so this entry sticks to what the page states. The canvas's code lives in a GitHub repository that is a mirror of the site's monorepo and has its own catalog entry; this entry covers the live canvas.
Questions about this entry
Short answers drawn from the linked project or page.
A shared canvas of a thousand by a thousand cells on chriswijnia.com, open to everyone. Pick an emoji and drag to paint, and Jev, TypeSafe AI's System One model, joins in by painting emoji of its own after your strokes.
Each visitor gets 300 cells of ink, with five coming back every second. The capture showed a full 300 of 300 ink bar, so the limit applies per visitor and refills over time rather than resetting on demand.
After each stroke it reads what is around it by name, weighs emoji that belong with it, and paints one. If it is sure that a shape was meant, it finishes it. The painting logic is part of the open-source code.
Yes. Everyone else's strokes arrive as they land, so the canvas is a shared surface rather than a private sandbox, and the activity feed shows recent paints. The capture shows the canvas connected to that shared state.
The canvas's rules and Jev's painter, in cwdx/1-million-emojis under the MIT licence. That GitHub repository is the hosted mirror of the site's code; this entry covers the live canvas at chriswijnia.com/lab/emoji.
The page includes a WebMCP tools section, indicating the canvas is exposed to agent tooling in browsers that support it. The capture does not detail the tool list or how an agent would authenticate.
In the lab description: Jev, TypeSafe AI's System One model, joins in after each stroke, and the page links the model name to TypeSafe's site. The palette and the painting behavior are described in the source repository.
More Jev resources like this
A semantic command palette demo where Jev ranks 77 commands by intent in Portuguese or English: one Choice question per keystroke, with the full probability distribution over the command catalog as the ranking. Runs side by side with a classic fuzzy matcher that fails honestly on the same queries.
Jev Demos & Experiments#community#catalog-import#cobanov
Community
amycardoso/jev-paletteJavaScript
amycardoso/jev-paletteJavaScriptA website companion's code: a page's actions are offered as WebMCP tools, a visitor's request becomes one Jev Choice, and a three.js voxel character hops to the button and does it.
Jev Demos & Experiments#community#catalog-import#cobanov
Community
cwdx/jevvieTypeScript
cwdx/jevvieTypeScriptA Hugging Face Space that runs a tiny System One model entirely in the browser: type a prompt and the app produces output on the spot, with no server connection or installation required.
Jev Demos & ExperimentsPlayground#community#catalog-import#cobanov
Community
JevSpeak explores whether a model that cannot generate text can still talk: about 13 parallel Jev decisions per turn become a typed Semantic IR that a deterministic compiler renders in English or Chinese. Live at jevspeak.org.
Jev Demos & Experiments#community#catalog-import#logicrw
Community
mm-sheng/jevspeakTypeScript
mm-sheng/jevspeakTypeScriptAn AI companion that replies only with facial expressions: Jev judges each message and picks one of 16 moods while blobatar morphs its face, with no text replies and no visible chat history.
Jev Demos & Experiments#community#catalog-import#logicrw
Community
appchainai/jevatarTypeScript
appchainai/jevatarTypeScriptA one-page playground and test bench for TypeSafe Jev: pick a backend, load an example and run typed questions against Jev through OpenRouter or TypeSafe, an LLM baseline, or a keyless mock — all speaking the same /v1/systemone contract, with an inspector for the answer distributions.
Jev Demos & Experiments#community#catalog-import#logicrw
Community
creativoma/here-we-go-jevTypeScript
creativoma/here-we-go-jevTypeScriptFrom the Jev guides
Original write-ups on what this entry is for.
From the Jev AI model community
Posts from builders shipping with Jev right now.