❗️

We have renamed CUSTOM mode to LITE mode to ease developer confusion. All functionality remains the same.

When using LiveAvatars in LITE Mode, we focus exclusively on real-time video generation driven by your audio input. Conversational orchestration—such as dialog management, reasoning, and response generation—remains fully controlled by your application.

LITe mode is recommended for developers that:

1. Already have an existing conversation stack or custom infrastructure in place.
   1. This includes proprietary or self-hosted LLMs, TTS pipelines, or dialogue systems.
2. Need finer-grained control over conversation flow, timing, or response logic than what FULL Mode provides
3. Want to integrate LiveAvatars as a modular video-rendering component within a broader, highly customized system.

Getting Started

Follow our quick start guide, where we walk through how to create a session token in LITE mode, and start the session. We recommend checking out our architectural guide for how we recommend setting up LITE mode.

In addition, we have some plugins built into several existing WebRTC frameworks such as LiveKit and Pipecat. Checkout our integrations to help quickly bootstrap your implementation.