End-to-End Integration

Integrate our end-to-end pipeline into your own apps.

In case you do not require control over the LLM or TTS model that is powering your application, we recommend using our end-to-end integration, which provides the lowest latency and most optimized multimodal experience of all integration options.

Our end-to-end agents are the fastest high-quality real-time avatars in the market, with response times as fast as 1s from user audio to HD avatar video.

Quickstart

Dashboard Flow

1. Create an Agent in the Dashboard

   Go to the Beyond Presence Dashboard, navigate to the Agents tab, click New Agent, and fill in the required information.

2. Copy Conversation Link

   Click on … > Share to copy the conversation link to talk to your agent.

API Flow

1. Create an Agent via the API

   Use the Create Agent API endpoint.

2. Retrieve the Agent ID

   Retrieve the id field from the agent creation response’s JSON body.

Creating an agent through the API is fast and flexible. It is ideal for generating just-in-time, single-use agents with dynamic, context-specific system prompts. This enables call handling that’s highly customized per interaction, which the dashboard flow does not currently support.

Embed your Agent

To embed your agent into your own website or product as an iFrame, use this snippet:

<iframe 
    src="https://bey.chat/defaultAgent"  // Replace with your own agent link or ID
    allow="camera; microphone" 
    allowfullscreen>
</iframe>

• Web users: Use the link copied from the dashboard’s Share menu.
• API users: Replace defaultAgent with your agent’s id.

Support for custom LLMs coming soon. In the meanwhile, we recommend to use the audio-to-video API instead if your use case requires a custom LLM. If you would like to use the end-to-end integration with custom LLM today, please reach out to us at [e-mail]!