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

# Models & Capabilities

> Choose a model, protocol, duration, resolution, and reference mode.

<AccordionGroup>
  <Accordion title="Which text protocols does FreeModel support?">
    FreeModel supports OpenAI Chat Completions, OpenAI Responses, Anthropic
    Messages, and Gemini native `generateContent` for the live GPT text routes.
  </Accordion>

  <Accordion title="Does FreeModel currently offer Claude text models?">
    No Claude model ID is in the current live catalog. Claude Code can connect
    through the Anthropic Messages-compatible protocol, but the selected model
    is a FreeModel GPT route.
  </Accordion>

  <Accordion title="Which Seedance 2 durations are available?">
    Both `video-ds-2.0` and `video-ds-2.0-fast` accept 5, 10, or 15 seconds.
    Prices are $1, $2, and \$3. Output is 720p.
  </Accordion>

  <Accordion title="Is familySlug part of the video API?">
    No. `familySlug` is an internal Portal value. Public video requests use
    `model`, `prompt`, and the selected model's media, duration, and framing
    fields.
  </Accordion>

  <Accordion title="Which Gemini Omni resolutions are available?">
    Choose `1280x720` or `1920x1080` for 16:9 output. Choose `720x1280` or
    `1080x1920` for 9:16 output. Gemini Omni does not support square output.
    T2V, I2V, and R2V generate a fixed 10-second result. Extend derives its
    duration from `video_url`, so do not send `seconds` for that model.
  </Accordion>

  <Accordion title="Which inputs do Gemini Omni video modes accept?">
    I2V requires one or two source images. R2V requires one to five reference
    images. Each image is limited to 5 MiB. Extend requires one source video in
    the singular `video_url` field and accepts up to five optional image
    references. T2V does not accept image or video inputs.
  </Accordion>

  <Accordion title="Which Veo 3.1 resolutions are available?">
    Model IDs without `-4k` produce 1080p. `veo3-1-lite-4k`, `veo3-1-fast-4k`,
    and `veo3-1-quality-4k` produce 4K. All Veo routes are fixed at 8 seconds.
  </Accordion>

  <Accordion title="How many Veo reference images can I send?">
    Lite, Fast, and Quality routes accept zero to two public image URLs.
    `veo3-1-fast-ref` requires one to three.
  </Accordion>

  <Accordion title="How do I discover newly added models?">
    Call [`GET /v1/models`](/api-reference/list-models) at runtime and check the
    [live models page](https://freemodel.app/models) for current prices.
  </Accordion>
</AccordionGroup>