# Announcing WebSocket Support with AsyncAPI v3

With full AsyncAPI support in Fern SDKs and Docs, you can now **document your WebSocket endpoints with ease.**  Fern continues to support AsyncAPI v2 and now includes full support for the newly released [v3](https://www.asyncapi.com/docs/migration/migrating-to-v3). Who benefits from low-latency, bi-directional APIs? It turns out Fintech (real-time market data) and Conversational AI (fast, dynamic interactions) are two hot areas.[Deepgram](https://deepgram.com/), a Voice AI platform for developers, is using AsyncAPI to deliver event-driven APIs to their customers.

![](/blog/67ba7e7a41e4436f8fb419d8_asyncapi-quote-320ba2d3.png)

## **How AsyncAPI v3 is built for modern event-driven architectures**

![](/blog/67ba7ea285a0230cb9b9b282_async-api-2-90eba020.png)

AsyncAPI v3 separates **operations** (like send and receive) from **channels**, making APIs more flexible and scalable. You can now reuse channels across services, define message flows more clearly, and handle multiple communication streams with ease. [Get all the details in the launch post](https://www.asyncapi.com/blog/release-notes-3.0.0).

## **Auto-generated code examples**

![](/blog/67ba7ec85741c3c70f3c1eb8_async-api-4-skewed-47fd8860.png)

Fern **auto-generates dynamic documentation** with real-world examples, showing you exactly how to send messages over WebSocket with AsyncAPI. No guesswork—just ready-to-use payloads and message formats that help you integrate faster and reduce errors. See them live in [Deepgram Docs](https://developers.deepgram.com/).

Get started with WebSockets in Fern today, [read the docs](https://buildwithfern.com/learn/docs/api-references/generate-websocket-ref).