# Avaluma AI

## Docs

- [Embed an Avaluma Agent on Your Website](https://docs.avaluma.ai/en/agents/embedding.md): Embed a shared Avaluma agent in any website with an iframe, and customize its behavior and appearance using query parameters.
- [Run an Avaluma Agent as a Fullscreen Kiosk](https://docs.avaluma.ai/en/agents/kiosk.md): Run a shared Avaluma agent fullscreen on an unattended display — kiosk stand, reception screen, or trade-show booth — with a screen wake lock and idle auto-reset.
- [Avaluma Agents: Create and Configure Avatars in the Dashboard](https://docs.avaluma.ai/en/agents/overview.md): Create, configure, and share photorealistic conversational avatars directly from the Avaluma dashboard at avaluma.ai — no infrastructure required.
- [Connect this documentation to your AI assistant via MCP](https://docs.avaluma.ai/en/mcp/connect.md): Query the Avaluma documentation directly from Claude Code, OpenAI Codex, Gemini CLI, and Google Antigravity using the hosted Model Context Protocol (MCP) server.
- [How Avaluma AI Works: Architecture Overview](https://docs.avaluma.ai/en/self-hosting/architecture.md): Explore the two-component pipeline that powers Avaluma AI: the GPU-rendered Avatar Server and the LiveKit Agent voice AI pipeline.
- [Avatar Server HTTPS: Configure a Caddy Reverse Proxy](https://docs.avaluma.ai/en/self-hosting/avatar-server/https-proxy.md): Add automatic TLS to your Avatar Server using the Caddy reverse proxy for production-ready HTTPS with automatic certificate management.
- [Avatar Server Overview: Deploy Photorealistic AI Avatars](https://docs.avaluma.ai/en/self-hosting/avatar-server/overview.md): Run the Avaluma Avatar Server to render .hvia avatars into live video and stream them into LiveKit rooms. Requires an NVIDIA GPU with 6+ GB VRAM.
- [Avatar Server Setup: Install and Configure with Docker](https://docs.avaluma.ai/en/self-hosting/avatar-server/setup.md): Deploy the Avaluma Avatar Server with Docker Compose, add your .hvia avatar files, and set the two required environment variables to go live.
- [Managing .hvia Avatar Files for the Avaluma Avatar Server](https://docs.avaluma.ai/en/self-hosting/configuration/avatar-files.md): Learn what .hvia avatar files are, how to obtain them, where to place them on disk, and how GPU VRAM limits affect simultaneous sessions.
- [Avaluma AI Environment Variables — Complete Reference](https://docs.avaluma.ai/en/self-hosting/configuration/environment-variables.md): A complete reference for every environment variable used by the Avaluma AI Avatar Server and LiveKit Agent, including required flags and defaults.
- [Avaluma AI: Photorealistic AI Avatars for Your App](https://docs.avaluma.ai/en/self-hosting/introduction.md): Avaluma AI lets you embed real-time photorealistic animated AI avatars into any application using GPU-accelerated rendering and LiveKit.
- [Build a Conversational Voice AI Agent with Avaluma](https://docs.avaluma.ai/en/self-hosting/livekit-agent/conversational-agent.md): Connect a full STT → LLM → TTS voice pipeline to an Avaluma avatar using LiveKit Inference, AssemblyAI, GPT-4.1-mini, and Cartesia Sonic-3.
- [Stream External Audio Directly to an Avaluma Avatar](https://docs.avaluma.ai/en/self-hosting/livekit-agent/external-audio.md): Send raw PCM audio directly to an Avaluma avatar via LiveKit DataStream, bypassing AgentSession — ideal for custom TTS or pre-generated audio sources.
- [LiveKit Agent: Connect a Voice AI Pipeline to Your Avatar](https://docs.avaluma.ai/en/self-hosting/livekit-agent/overview.md): Drive an Avaluma avatar with a Python LiveKit Agent. Choose between a full voice AI pipeline or direct external audio streaming via DataStream.
- [Get Started with Avaluma AI Avatars](https://docs.avaluma.ai/en/self-hosting/quickstart.md): Deploy the Avatar Server and LiveKit Agent with Docker, then connect a test client to see your photorealistic AI avatar live in under 15 minutes.
- [Connect a Test Client to Your Avaluma AI Avatar Agent](https://docs.avaluma.ai/en/self-hosting/testing/test-clients.md): Use the Avaluma Test Client or LiveKit Agent Playground to verify your avatar pipeline end-to-end — choose the right client for your LiveKit setup.
- [Troubleshoot Avaluma AI Sessions, GPU, and Agent Issues](https://docs.avaluma.ai/en/self-hosting/testing/troubleshooting.md): Diagnose and fix the most common issues with Avaluma AI avatar sessions, GPU errors, connectivity problems, and LiveKit agent configuration.