Skip to content
Home

Nyxclaw

Gives AI agents a real-time face and voice, with two voice pipelines (OpenAI Realtime and local CPU) and 52 ARKit blendshapes at 30 FPS.

This open-source software provides real-time AI avatar companions with face and voice for AI agents, solving the problem of giving agents a human-like interface. It targets individual developers and consumers who want to customize their AI agents, running on their own machines without needing a GPU. The product is self-hosted, ensuring data privacy and security, and offers both cloud-based and local voice pipelines for flexibility.

Key features

  • Real-time face and voice for AI agents
  • Runs on CPU, no GPU required
  • Open source and free
  • Self-hosted server with Docker or install script
  • End-to-end encrypted with Ed25519 auth
  • QR pairing for device connection
  • Two voice pipelines: OpenAI Realtime and local CPU
  • Wav2Arkit ONNX for 52 ARKit blendshapes at 30 FPS
Social posts
  • No social media activity within the last 30 days
GTM channels
  • No GTM activity detected
ICP
  • Software developers
  • Consumers general
  • Freelancers solopreneurs