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

# The canvas

> Add nodes, connect them into a workflow, pick a model, and run to generate creative.

The canvas is a node-based workspace. You add **nodes**, connect their outputs to other nodes' inputs, and run the graph to generate or edit media. Connections are typed, so the canvas only lets you make links that make sense.

## Building a graph

* **Right-click the canvas** to open the add menu and spawn a node (Image, Video, or Utility). Nodes appear where you clicked.
* **Drag a file** from your computer or the media library onto the canvas to create a reference node.
* **Connect nodes** by dragging from one node's output handle to another's input.
* **Right-click a node** for its actions — duplicate, delete, run, and configuration.

## Nodes

Nodes are grouped in the library exactly as they appear on the palette.

### Image

| Node                 | What it does                                   |
| -------------------- | ---------------------------------------------- |
| **Image Generation** | Generate images with a selectable image model. |
| **Image Reference**  | Bring an image in as a reference input.        |

### Video

| Node                  | What it does                                                |
| --------------------- | ----------------------------------------------------------- |
| **Video Generation**  | Generate clips with a selectable video model.               |
| **Omni Flash (Edit)** | Generate a clip, then chat to edit it into variations.      |
| **Extend Video**      | Continue existing footage.                                  |
| **Video Splicer**     | Concatenate clips in the browser.                           |
| **Video Editor**      | A full editor — trim, split, and sequence clips and stills. |
| **Video Reference**   | Bring a video in as a reference input.                      |
| **Video Decoder**     | Break a video down frame by frame for creative analysis.    |

### Utility

| Node                   | What it does                                                                     |
| ---------------------- | -------------------------------------------------------------------------------- |
| **Audio Reference**    | Bring in voice or sound.                                                         |
| **Document Context**   | Add PDF or text knowledge to a generation.                                       |
| **Text Block**         | A prompt or enrichment input.                                                    |
| **Publish to Planner** | Attach an edited video to an organic draft in the [calendar](/organic/calendar). |
| **Note**               | A free-text annotation on the canvas.                                            |

## Models

Generation nodes let you choose a model to match the job:

| Type      | Models                                      |
| --------- | ------------------------------------------- |
| **Image** | Nano Banana, Nano Banana Pro, Nano Banana 2 |
| **Video** | Veo 3.1, Veo 3.1 Fast, Veo 3.1 Lite, Kling  |

Aspect ratios and other options are validated per model, so the canvas only offers settings a model actually supports.

## Running a graph

A generation node runs once every input it depends on has valid data. You can run a single node or let a connected chain run in order. Each node shows its status:

* `idle` — waiting or not yet run
* `running` — generating
* `success` — output ready
* `error` — something went wrong

Outputs are saved to your [Library](/studio/library), and you can download any generated artifact.

## Keeping generations on-brand

Generation nodes are grounded in your [brand book](/brand/overview) by default, and you can apply [creative skills](/studio/workflows) per node. Both are covered in [Workflows & skills](/studio/workflows).

## Shortcuts

| Action          | Shortcut            |
| --------------- | ------------------- |
| Pan (hand tool) | `H`                 |
| Select          | `V`                 |
| Fit view        | `Shift+F`           |
| Duplicate node  | `Cmd+D`             |
| Run node        | `R`                 |
| Delete node     | `Del` / `Backspace` |

## Collaboration

The canvas is multiplayer. When teammates are in the same canvas, you see their cursors and which nodes they have selected, updating in real time.

## What's next

<CardGroup cols={2}>
  <Card title="Workflows & skills" icon="diagram-project" href="/studio/workflows">
    Save and reuse graphs; apply creative skills and brand grounding.
  </Card>

  <Card title="Library" icon="images" href="/studio/library">
    Find your generated creative and references.
  </Card>
</CardGroup>