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

# Limits & constraints

> Request limits, output format, rate limits, and cold starts.

## Request limits

| Limit                   | Value                                    |
| ----------------------- | ---------------------------------------- |
| Text length             | up to **\~5,000 characters** per request |
| `nfe` (diffusion steps) | `8`–`64` (`16` fast, `32` quality)       |
| `speed`                 | `0.5`–`2.0`                              |

For longer inputs, split on sentence/paragraph boundaries and synthesize in parallel, or use [`tts.stream()`](/sdk/typescript) which chunks for you.

## Output format

* **`tts.create` / `/v1/tts`:** WAV — **24 kHz, 16-bit PCM, mono**.
* **`/v1/audio/speech`:** `wav` or `pcm` (raw 16-bit PCM @ 24 kHz).
* **`tts.stream` / `/v1/tts/stream`:** a sequence of WAV chunks (framed).
* MP3 / Opus / µ-law are not native yet — [convert with `ffmpeg`](/guides/playing-audio) ([roadmap](/roadmap)).

## Models & voices

* **Model:** `tokay-kk-v1` (Kazakh) today. New models/languages appear on `GET /v1/models`.
* **Voices:** six for `tokay-kk-v1` (`Aigerim` default). Always fetch [`/v1/voices`](/voices) rather than hard-coding.

## Rate limits

There is **no hard per-second rate limit** today — throughput is bounded by your [credit balance](/billing). Requests run concurrently; the SDK does not serialize them. (Explicit rate limiting may arrive later.)

## Cold starts

The GPU backend **scales to zero** when idle to save cost, so the **first request after an idle period** can take tens of seconds; subsequent calls (within a few minutes) are fast. The SDK's default **120 s** timeout is sized for this — see [Latency & quality](/guides/latency) for tuning and keep-warm tips.

## Next steps

<CardGroup cols={2}>
  <Card title="Latency & quality" icon="gauge-high" href="/guides/latency">
    Tune the `nfe` dial and handle cold starts.
  </Card>

  <Card title="Streaming" icon="waveform-lines" href="/guides/streaming">
    Chunk long inputs to start playback sooner.
  </Card>

  <Card title="Billing & credits" icon="credit-card" href="/billing">
    Throughput is bounded by your credit balance.
  </Card>

  <Card title="Errors & retries" icon="shield-check" href="/sdk/errors">
    What happens when a request exceeds a limit.
  </Card>
</CardGroup>