AllPersonas for developers · Alpha
Face for your AI agent
Bring your agent’s replies to life with expressive faces and speech. Use your own model, voice, and web framework.
MIT code and artwork · No API key needed to render a face.
6 characters. 24 expressions. One SDK. Choose a face to try it.
Built for developers
A few lines of code.
A face of its own.
Your app supplies the conversation. AllPersonas handles the face, expressions, and speech playback. Start with a working example, then connect it to your agent.
Read the quickstartimport { createFace } from '@allpersonas/core';
export function mountAssistant(host: HTMLElement) {
host.style.width = '280px';
host.style.height = '340px';
const face = createFace(host, {
avatar: 'ruby',
expression: 'attentive',
});
face.update({ thinking: true });
// When your agent responds:
face.update({ thinking: false, expression: 'joyful' });
// Call the returned function when your view unmounts.
return () => face.destroy();
}Run the starter
Try a complete local app with expressions, browser speech, and Stop controls. Demo mode uses sample replies and needs no API key.
Connect your model
Use your existing agent or try OpenAI, Anthropic, or Vercel AI Gateway in the playground with your own key.
Choose your voice
Use browser speech, connect ElevenLabs in the playground, or drive the face from your app’s audio.
Stay in the loop
Updates from AllPersonas
New characters and SDK releases, in your inbox.
Follow new characters and SDK releases.
The SDK is available now. Sign up for release updates.