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

> ## Agent Instructions
> The Deepshi API is an OpenAI-compatible gateway. Base URL: https://api.deepshi.ai/v1. API keys start with sk-bf- and go in the Authorization: Bearer header. Prefer the official OpenAI SDKs pointed at the Deepshi base URL. Model ids are clean with no provider prefix (e.g. deepshi-3.0, claude-opus-4.8, gpt-5.5). Chat and image requests are synchronous; video and music requests are asynchronous job APIs (create, then poll). Every synchronous response carries usage.cost.total_cost in USD. Do not reference the internal /api/* admin plane or virtual keys.

# Audio models

> Audio models available through the Deepshi API, with prompt limits and output options.

Use these models with the [audio generation](/capabilities/audio-generation) endpoints. Reference a model by its **id** in the `model` field.

<Note>
  The exact set your key can call, and each model's supported options, is returned by `GET /v1/models`.
  Audio is priced per track or per minute of output depending on the model; for per-model rates see
  [deepshi.ai](https://deepshi.ai/).
</Note>

Recently added models appear first.

<CardGroup cols={2}>
  <Card title="Lyria 3 Pro" icon="https://mintcdn.com/evermindlabs/DGZ60wKm5MW0do78/logos/google.svg?fit=max&auto=format&n=DGZ60wKm5MW0do78&q=85&s=10fdc070e84e6698933120255ba20e4d" width="16" height="16" data-path="logos/google.svg">
    `lyria3-pro` · **Prompt** 5K

    The full-length Lyria tier: a complete multi-minute vocal song from a single prompt.

    <Icon icon="microphone" /> Vocals
  </Card>

  <Card title="MiniMax Music 2.5" icon="https://mintcdn.com/evermindlabs/DGZ60wKm5MW0do78/logos/minimax.svg?fit=max&auto=format&n=DGZ60wKm5MW0do78&q=85&s=b9c2dc9a9285c81193fd647e9ecca1a5" width="16" height="16" data-path="logos/minimax.svg">
    `minimax-music-v2.5` · **Prompt** 2K

    Vocal songs or instrumentals, with optional lyrics.

    <Icon icon="microphone" /> Vocals  <Icon icon="music" /> Instrumental
  </Card>

  <Card title="MiniMax Music 2.6" icon="https://mintcdn.com/evermindlabs/DGZ60wKm5MW0do78/logos/minimax.svg?fit=max&auto=format&n=DGZ60wKm5MW0do78&q=85&s=b9c2dc9a9285c81193fd647e9ecca1a5" width="16" height="16" data-path="logos/minimax.svg">
    `minimax-music-v2.6` · **Prompt** 2K

    The latest MiniMax tier, with the same controls as 2.5.

    <Icon icon="microphone" /> Vocals  <Icon icon="music" /> Instrumental
  </Card>

  <Card title="Lyria 3" icon="https://mintcdn.com/evermindlabs/DGZ60wKm5MW0do78/logos/google.svg?fit=max&auto=format&n=DGZ60wKm5MW0do78&q=85&s=10fdc070e84e6698933120255ba20e4d" width="16" height="16" data-path="logos/google.svg">
    `lyria3` · **Prompt** 2K

    Google's Lyria 3 generates a multilingual vocal song from a single prompt, \~30 seconds.

    <Icon icon="microphone" /> Vocals
  </Card>

  <Card title="ElevenLabs Music" icon="https://mintcdn.com/evermindlabs/DGZ60wKm5MW0do78/logos/elevenlabs.svg?fit=max&auto=format&n=DGZ60wKm5MW0do78&q=85&s=ce25c8c57fef5a93286a237dc3fa2640" width="16" height="16" data-path="logos/elevenlabs.svg">
    `elevenlabs-music` · **Prompt** 2K

    A song from a style prompt, with controllable length and optional vocals.

    <Icon icon="microphone" /> Vocals  <Icon icon="music" /> Instrumental  <Icon icon="clock" /> Duration
  </Card>

  <Card title="Stable Audio 2.5" icon="https://mintcdn.com/evermindlabs/DGZ60wKm5MW0do78/logos/stability.svg?fit=max&auto=format&n=DGZ60wKm5MW0do78&q=85&s=1932a395f0f89d658ab4767a73215c4e" width="16" height="16" data-path="logos/stability.svg">
    `stable-audio-25` · **Prompt** 2K

    Stability's instrumental music and sound effects, with controllable length.

    <Icon icon="music" /> Instrumental  <Icon icon="clock" /> Duration
  </Card>

  <Card title="CassetteAI Music Generator" icon="https://mintcdn.com/evermindlabs/DGZ60wKm5MW0do78/logos/cassetteai.svg?fit=max&auto=format&n=DGZ60wKm5MW0do78&q=85&s=476d2d78fc4fa046ffab44b63557a956" width="16" height="16" data-path="logos/cassetteai.svg">
    `cassetteai-music` · **Prompt** 2K

    CassetteAI's instrumental music, from 10 to 180 seconds.

    <Icon icon="music" /> Instrumental  <Icon icon="clock" /> Duration
  </Card>

  <Card title="MiniMax Music 2.0" icon="https://mintcdn.com/evermindlabs/DGZ60wKm5MW0do78/logos/minimax.svg?fit=max&auto=format&n=DGZ60wKm5MW0do78&q=85&s=b9c2dc9a9285c81193fd647e9ecca1a5" width="16" height="16" data-path="logos/minimax.svg">
    `minimax-music-v2` · **Prompt** 300

    Studio-quality vocal songs from a style prompt and required lyrics.

    <Icon icon="microphone" /> Vocals
  </Card>
</CardGroup>

<Tip>
  Not sure where to start? **`minimax-music-v2.6`** is a strong vocal-song default, and
  **`stable-audio-25`** and **`cassetteai-music`** are good instrumental options that skip lyrics.
</Tip>