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

# Omni Video Models Overview

> Documentation for the `omni-fast` and `omni-fast-v2v` video generation APIs, organized by model.

# Omni Video Models

The Omni video models submit asynchronous tasks via `POST /v1/videos` and check status with `GET /v1/videos/{task_id}`.

<CardGroup cols={2}>
  <Card title="omni-fast" icon="bolt" href="/en/videos/omni/omni-fast">
    Text-to-video, first-frame image-to-video, first/last frame, and multi-reference image video.
  </Card>

  <Card title="omni-fast-v2v" icon="film" href="/en/videos/omni/omni-fast-v2v">
    Reference-video editing, extension, and regeneration.
  </Card>
</CardGroup>

## Supported Models

| Model           | Description              | Supported Modes                                                            | Reference Asset Limits                          |
| --------------- | ------------------------ | -------------------------------------------------------------------------- | ----------------------------------------------- |
| `omni-fast`     | General fast video model | Text-to-video, first-frame I2V, first/last frame, multi-image              | Up to 5 `images`; reference video not supported |
| `omni-fast-v2v` | Reference-video model    | Reference-video editing, extension, regeneration; partial T2V / I2V fields | Reference video up to 15MB                      |

## Supported Capabilities

| Capability       | `omni-fast` | `omni-fast-v2v` | Key Fields                                   |
| ---------------- | ----------- | --------------- | -------------------------------------------- |
| Text-to-video    | Yes         | Yes             | `model`, `prompt`, `seconds`, `aspect_ratio` |
| First-frame I2V  | Yes         | Yes             | `first_image_url`                            |
| First/last frame | Yes         | Yes             | `first_image_url`, `last_image_url`          |
| Multi-reference  | Yes         | Yes             | `images`                                     |
| Reference video  | No          | Yes             | `video` / `video_url` / `input_video`        |

## Unified Endpoint

```http theme={null}
POST /v1/videos
```

All Omni video tasks use a JSON request body. On success, an OpenAI-style `video` object is returned, where `id` or `task_id` is used for subsequent queries.

## Task Query

```http theme={null}
GET /v1/videos/{task_id}
```

Use `GET /v1/videos/{task_id}` to poll the status. After the task completes, retrieve the generated MP4 from the returned `video_url` (or `data[0].url`).

## Output Limits

| Item             | Rule                                                            |
| ---------------- | --------------------------------------------------------------- |
| Video duration   | `seconds` is recommended as a string, range `4` to `30` seconds |
| Resolution       | `resolution` example values: `720p`, `1080p`, `2k`, `4k`        |
| Aspect ratio     | `aspect_ratio` example values: `16:9`, `9:16`                   |
| Reference images | Public URL or `data:image/...;base64,...`; up to 5 `images`     |
| Reference video  | Only supported by `omni-fast-v2v`, max 15MB                     |

## Sub-API Pages

| Page                             | Content                                                      |
| -------------------------------- | ------------------------------------------------------------ |
| [omni-fast](./omni-fast)         | Text-to-video, image-to-video, first/last frame, multi-image |
| [omni-fast-v2v](./omni-fast-v2v) | Reference-video editing, extension, regeneration             |

## Recommended Reading Order

1. [omni-fast](./omni-fast)
2. [omni-fast-v2v](./omni-fast-v2v)
3. [Video Model Support Matrix](/en/videos/model-matrix)
4. [Task Status API](/en/tasks/task-status)

## Related Pages

* [Video Model Support Matrix](/en/videos/model-matrix)
* [OpenAI Video Compatible API](/en/videos/openai-videos)
* [Task Status API](/en/tasks/task-status)