What is the Orga AI API?

The Orga AI API enables real-time interaction with our AI model Orga, allowing your applications to stream user video+audio or audio-only while receiving low-latency audio responses. Powered by WebRTC for bidirectional media and WebSocket/HTTPS for control messages (authentication, parameters, events), it simplifies building AI-driven conversations. Developers can use it directly with a backend proxy or through our SDKs for streamlined integration.

Why use Orga AI?

Benefit	What it Means for You
Real-time AI	Enables natural, interactive conversations with instant AI audio replies.
Multimodal Input	Stream user video or audio—show objects, share screens, or speak naturally.
Audio Output	Receives clear AI audio responses, optimized for voice interactions.
Developer Ease	SDKs and a secure proxy simplify setup and protect API keys.

Use Cases

Scenario	3-Line Gist
Voice Bot	1️⃣ Capture microphone audio. 2️⃣ POST to /v1/realtime/calls. 3️⃣ Play AI audio reply.
Video Assistant	1️⃣ Stream camera video and audio. 2️⃣ POST to /v1/realtime/calls. 3️⃣ Hear AI audio response.
Screen-Share Tutor	1️⃣ Share screen video and audio. 2️⃣ POST to /v1/realtime/calls. 3️⃣ Receive AI spoken guidance.

Explore implementation details in the cURL Quick Start, React Native SDK, or React SDK for SDK-based solutions, or build custom integrations via the API Reference.

cURL Quickstart

This guide helps developers test and set up the Orga AI API for real-time WebRTC sessions by fetching client secrets and ICE servers using cURL. Use the cURL scripts to test /v1/realtime/client-secrets and /v1/realtime/ice-config or adapt them into a backend proxy for secure integration with the /v1/realtime/calls endpoint, which requires a WebRTC-capable environment (browser or mobile app).

React Native SDK

Link to a page in the guide

Web SDK

Link to a page in the guide