Your live AI persona, in the browser.
Chat with a 3D avatar that listens in real-time and speaks back in your voice of choice. Free to try, no card required.
Sub-second token streaming. Choose from open-weights models on free, frontier models (Claude, GPT, Qwen) on pro.
Push-to-talk in browser, Whisper-large-v3 transcription, ElevenLabs / Kokoro / Cartesia voice synthesis with lipsync.
VRoid-style VRM models render in real time with viseme lipsync and idle motion. Bring your own .vrm URL on pro.
Conversations encrypted at rest with AES-256-GCM. We never train on your content. No analytics, no ad trackers.
Free is genuinely free — no card required, no time limit. Upgrade when you want premium voices and frontier chat models.