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

# Grounded Generation

> Generate images and video grounded in real web references.

<p style={{ fontSize: "0.8rem", opacity: 0.6 }}>For AI agents: <a href="https://docs.octen.ai/capabilities/grounded-generation.md">docs.octen.ai/capabilities/grounded-generation.md</a></p>

Generation models create from training data, and anything outside it suffers. Something real, such as a product, a place, or an event, drifts from the facts. Anything that appeared after training is simply unknown to the model. An unfamiliar subject or a specific action comes out invented rather than accurate.

Octen Grounded Generation searches first: it retrieves the images and videos that show the subject, then generates an image or a video grounded in that material. With real references to work from, the output is simply better.

For the full list of parameters, see the [Grounded Generation API reference](/api-reference/post-grounded-generation).

## Why Grounded Generation

* **Faithful to the real thing.** Generation is anchored to retrieved web references, so real subjects look like themselves.
* **Current by default.** References come from live search, so generated visuals reflect how things look now, not at training time.
* **One step from prompt to visual.** Search and generation run as a single workflow, with no retrieval pipeline to build on your side.

## How It Works

1. Send a generation prompt and say whether you want an image or a video. The call returns a task id right away.
2. Octen searches the web for the images and videos that show the subject, then picks the references to generate from.
3. A generation model creates the output from those references. Poll the task until it finishes, then download the result from the returned link.

## Scenarios

### Visuals of real subjects

Generate products, places, people, and events from live web references instead of the model's memory.

```json theme={null}
{
  "modality": "image",
  "prompt": "Ronaldo lifting the World Cup trophy, stadium lights",
  "aspect_ratio": "16:9"
}
```

### Illustrate an article or answer

Turn a topic into a supporting video grounded in what the web actually shows.

```json theme={null}
{
  "modality": "video",
  "prompt": "A Tesla Cybertruck driving through a snowy mountain pass at dawn"
}
```

### Vertical video for social feeds

Match the shape and length a feed expects.

```json theme={null}
{
  "modality": "video",
  "prompt": "Barista pouring latte art in a sunlit cafe",
  "aspect_ratio": "9:16",
  "duration": 10
}
```

## Next Steps

<CardGroup cols={2}>
  <Card title="Try in Console" icon="play" href="https://octen.ai/platform/grounded-generation">
    Run Grounded Generation live in the Octen console.
  </Card>

  <Card title="Grounded Generation API Reference" icon="code" href="/api-reference/post-grounded-generation">
    Full request/response schema.
  </Card>
</CardGroup>

***

<p style={{ fontSize: "0.8rem", opacity: 0.6 }}>Last modified on October 9, 2026</p>


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.