# Eleven v4 Turbo

> Eleven v4 Turbo is the real-time variant of the ElevenLabs Eleven v4 model family, with median inference latency of approximately 100 ms. It became available on September 28, 2026. It is used through the Text to Dialogue WebSocket with the API id eleven_v4_turbo, for agents and interactive applications.

URL: https://aistack.sh/model/eleven-v4-turbo
Vendor: ElevenLabs (product page: https://aistack.sh/tool/elevenlabs)
Last checked against sources: 2026-09-30

## Specs

- Released: Sep 28, 2026
- Status: Available
- API model ID: eleven_v4_turbo

## Where you can use it

- Eleven v4 Turbo is the real-time variant of the Eleven v4 family, with median inference latency of approximately 100 ms.
- For content creation and long-form audio, ElevenLabs points to eleven_v4 through the Text to Dialogue API instead.

## FAQ

### What is the API model ID for Eleven v4 Turbo?

eleven_v4_turbo, used through the Text to Dialogue WebSocket, according to the ElevenLabs changelog.

### How fast is Eleven v4 Turbo?

ElevenLabs says its median inference latency is approximately 100 ms.

### When did Eleven v4 Turbo become available?

It was announced as available in the ElevenLabs changelog entry dated September 28, 2026.

### How is Eleven v4 Turbo different from Eleven v4?

Both are in the same model family. ElevenLabs recommends eleven_v4_turbo over the Text to Dialogue WebSocket for agents and interactive applications, and eleven_v4 through the Text to Dialogue API for content creation and long-form audio.

## Sources

- [ElevenLabs: Changelog](https://elevenlabs.io/docs/changelog), read 2026-09-30