> ## Documentation Index
> Fetch the complete documentation index at: https://docs.cartesia.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Ink 2

Ink 2 is our fastest, most accurate streaming speech-to-text model — the lowest word error rate and best turn detection of any streaming STT, built for production voice agents. It transcribes structured data like phone numbers, dates, and emails right the first time, and knows when a speaker starts and finishes — no separate VAD required.

Turn detection is built in. Ink 2 emits a full lifecycle of turn events — `turn.start`, `turn.update`, `turn.eager_end`, `turn.resume`, and `turn.end` — so your agent knows exactly when to listen, think, and respond. See [Turn Events](/use-the-api/stt/turns) for the state machine, or [Compare STT Endpoints](/use-the-api/stt/compare-endpoints) to run Ink 2 with or without turn detection.

| Model   | Release Date | Languages | Status                              |
| ------- | ------------ | --------- | ----------------------------------- |
| `ink-2` | May 22, 2026 | `en`      | <Badge color="green">Stable</Badge> |

For information on `ink-whisper`, see our page on [Older STT Models](/build-with-cartesia/stt/older-models).

## Where to go next

<CardGroup cols={2}>
  <Card title="Try it out online" icon="arrow-pointer" href="https://www.cartesia.ai/ink">
    No sign-up or code required
  </Card>

  <Card title="Start building" icon="readme" href="/build-with-cartesia/stt/guides">
    Guides and best practices
  </Card>
</CardGroup>