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

# Videos

> Generate videos from text prompts with the praisonai videos command

The `praisonai videos` command turns a text prompt into a video from the command line.

```mermaid theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}}
graph LR
    subgraph "praisonai videos"
        Prompt[📝 Prompt] --> CLI[🎬 videos]
        CLI --> Video[✅ MP4]
    end

    classDef input fill:#6366F1,stroke:#7C90A0,color:#fff
    classDef process fill:#189AB4,stroke:#7C90A0,color:#fff
    classDef output fill:#10B981,stroke:#7C90A0,color:#fff

    class Prompt input
    class CLI process
    class Video output
```

## Quick Start

<Steps>
  <Step title="Generate a video">
    Omit `--model` to use the canonical default `openai/sora-2`.

    ```bash theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}}
    praisonai videos "A cat playing with yarn"
    ```
  </Step>

  <Step title="Save to a file">
    ```bash theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}}
    praisonai videos "A sunset over the ocean" --output sunset.mp4
    ```
  </Step>
</Steps>

***

## How It Works

The command forwards your prompt to the video generation capability and prints the resulting video ID, status, and URL.

| Step   | What happens                                          |
| ------ | ----------------------------------------------------- |
| Prompt | You describe the video in plain text                  |
| Model  | `--model` selects the route (default `openai/sora-2`) |
| Output | `--output` optionally saves the video to disk         |

***

## Options

| Option     | Short | Default         | Description                     |
| ---------- | ----- | --------------- | ------------------------------- |
| `--model`  | `-m`  | `openai/sora-2` | Provider-prefixed LiteLLM route |
| `--output` | `-o`  | `None`          | Output file path                |

Only `--model` and `--output` are configurable on the CLI today.

***

## Common Patterns

Pick a specific model:

```bash theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}}
praisonai videos "A city skyline at night" -m openai/sora-2-pro
```

Save the video to disk:

```bash theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}}
praisonai videos "A waterfall in a forest" -o waterfall.mp4
```

***

## Best Practices

<AccordionGroup>
  <Accordion title="Rely on the default model">
    Omit `--model` to use `openai/sora-2`, the canonical default shared by every video surface. Set `--model` only when you need a different provider route.
  </Accordion>

  <Accordion title="Use provider-prefixed routes">
    Always pass a full LiteLLM route such as `openai/sora-2`. Bare names like `sora` are not valid routes and cause routing errors.
  </Accordion>

  <Accordion title="Save long videos to disk">
    Use `--output` to write the result to a file so you keep a local copy of the generated video.
  </Accordion>
</AccordionGroup>

***

## Related

<CardGroup cols={2}>
  <Card title="Videos Capability" icon="video" href="/docs/capabilities/videos">
    The Python `video_generate` API.
  </Card>

  <Card title="CLI Reference" icon="terminal" href="/docs/cli/cli-reference">
    All PraisonAI CLI commands.
  </Card>
</CardGroup>