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

# Send provider-specific video options

> Add provider-specific video settings through `provider_params` without losing the shared gateway request shape.

Use this recipe when the common video request fields are not enough and one provider path needs extra settings.

## Goal

* Keep the shared gateway request shape.
* Add provider-specific overrides only where required.
* Avoid forking your whole integration around one provider.

## 1. Start with the shared video request

Keep the common request fields at the top level:

* `model`
* `prompt`
* `duration`
* `aspect_ratio`
* `resolution`
* `input_references`
* `webhook`

Only use `provider_params` for the extra provider-specific fields.

## 2. Add provider-specific settings under `provider_params`

<CodeGroup>
  ```bash cURL theme={null}
  curl https://api.phaseo.app/v1/videos \
    -H "Authorization: Bearer YOUR_API_KEY" \
    -H "Content-Type: application/json" \
    -d '{
      "model": "openai/sora-2",
      "prompt": "A clean product reveal with a slow dolly forward.",
      "duration": 8,
      "provider_params": {
        "camera_motion": "dolly_in",
        "style_preset": "product_clean"
      }
    }'
  ```

  ```typescript TypeScript theme={null}
  const response = await fetch("https://api.phaseo.app/v1/videos", {
    method: "POST",
    headers: {
      Authorization: "Bearer YOUR_API_KEY",
      "Content-Type": "application/json",
    },
    body: JSON.stringify({
      model: "openai/sora-2",
      prompt: "A clean product reveal with a slow dolly forward.",
      duration: 8,
      provider_params: {
        camera_motion: "dolly_in",
        style_preset: "product_clean",
      },
    }),
  });

  const job = await response.json();
  console.log(job.video_id);
  ```

  ```python Python theme={null}
  import requests

  response = requests.post(
      "https://api.phaseo.app/v1/videos",
      headers={
          "Authorization": "Bearer YOUR_API_KEY",
          "Content-Type": "application/json",
      },
      json={
          "model": "openai/sora-2",
          "prompt": "A clean product reveal with a slow dolly forward.",
          "duration": 8,
          "provider_params": {
              "camera_motion": "dolly_in",
              "style_preset": "product_clean",
          },
      },
  )

  job = response.json()
  print(job.get("video_id"))
  ```
</CodeGroup>

The exact accepted keys depend on the provider and model path. Keep provider documentation as the source of truth for the enum values and option names that sit inside this object.

## 3. Keep overrides isolated

Do not duplicate the whole request just because one provider needs two extra fields.

Good pattern:

* one gateway request shape
* one optional `provider_params` object

Bad pattern:

* one code path per provider for otherwise identical video jobs

## 4. Validate with one known model path first

Before generalizing the override:

* choose one known model
* submit one real job
* confirm the status and final output

Provider-specific video controls are exactly where silent request mismatches become expensive if you try to abstract too early.

## 5. Document the override at the call site

Whenever you add `provider_params`, leave a short comment or internal note that explains:

* which provider path requires it
* why the top-level gateway fields were not enough
* where the provider docs live

That keeps the exception maintainable later.

## Related guides

* [Use image inputs for video generation](./use-image-inputs-for-video-generation.mdx)
* [Run async video with webhooks](./async-video-webhooks.mdx)
* [API Reference: Video generation](../api-reference/endpoint/video-generation.mdx)