> ## 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.

# FAQ

> Common questions about the Deepshi API.

<AccordionGroup>
  <Accordion title="Is the API OpenAI-compatible?">
    Yes. The base URL is `https://api.deepshi.ai/v1` and the request/response
    shapes match OpenAI's, so the official OpenAI SDKs and most compatible tools
    work after changing the base URL and key. See
    [Migrating from OpenAI](/guides/migrating-from-openai).
  </Accordion>

  <Accordion title="How do I get an API key?">
    Create one from your [Deepshi dashboard](https://deepshi.ai/). Keys
    are shown only once. See
    [Authentication](/get-started/authentication).
  </Accordion>

  <Accordion title="Which models can I use?">
    Deepshi's own models plus leading third-party text, image, video, and music
    models from OpenAI, Anthropic, Google, xAI, and others. The live list your
    key can call is at `GET /v1/models`. See [Models](/models/overview).
  </Accordion>

  <Accordion title="How am I billed?">
    The API is prepaid with USD credits shared across all your keys. Usage is
    priced per model. See [Credits & billing](/get-started/pricing-credits).
  </Accordion>

  <Accordion title="What happens when I run out of credits?">
    Inference requests return `402 Payment Required` until you top up your
    balance from the dashboard.
  </Accordion>

  <Accordion title="Do API keys expire?">
    No. A key stays valid until you rotate, revoke, or delete it from the
    dashboard.
  </Accordion>

  <Accordion title="Are my prompts used for training?">
    No. Your prompts and completions are not used to train models. See
    [Privacy](/overview/privacy) for details.
  </Accordion>

  <Accordion title="Do you store or log my prompts?">
    No. Your prompt and response content is never stored, on Deepshi's own
    infrastructure or with any provider we route to, and we hold zero-retention
    agreements with those providers. See [Privacy](/overview/privacy).
  </Accordion>

  <Accordion title="How do I get my generated video and music files?">
    A completed job returns a direct download URL. Save the file as soon as the
    job completes. See [video generation](/capabilities/video-generation) and
    [music generation](/capabilities/audio-generation).
  </Accordion>

  <Accordion title="Am I charged for a failed generation?">
    No. A failed request (4xx) and a failed video or music job are not billed.
    For video, canceling a job before it completes also means it is not charged.
  </Accordion>

  <Accordion title="How do I get help?">
    Email [support@deepshi.ai](mailto:support@deepshi.ai).
  </Accordion>
</AccordionGroup>