# Cinematic Avatar

> Make a short cinematic clip with 1 to 3 HeyGen avatar looks from a text prompt. No script or voice; the prompt directs the scene, action and camera.

Source: https://nodaro.ai/docs/nodes/video/cinematic-avatar

The **Cinematic Avatar** node makes a short cinematic clip that features 1 to 3 HeyGen avatar looks. There is no script and no voice: a text prompt describes the scene, the action and the direction, and HeyGen renders the clip with the looks you chose.

- Found in: Video › Animate & Perform
- Output: video
- API type: `cinematic-avatar`

## When to use it
- A cinematic intro, a teaser or a stylized moment for a presenter you already use in [AI Avatar](https://nodaro.ai/docs/nodes/video/ai-avatar).
- A story beat with a consistent set of avatar looks.
- A quick generative clip that needs no lip-synced narration.

For a talking-head video with a script or a voiceover, use [AI Avatar](https://nodaro.ai/docs/nodes/video/ai-avatar) instead.

## Quick start
### Add the node

Press Tab on the canvas and choose **Video › Animate & Perform › Cinematic Avatar**.

### Pick the avatar looks

Select the node. Under **Avatar Looks**, pick 1 to 3 looks from the searchable avatar catalog, with a preview of each.

### Describe the scene

Write the **Prompt**: the scene, the mood, the camera and the action. Or connect a text node to the **Prompt** input.

### Set the clip and run it

Choose the **Duration**, **Aspect Ratio** and **Resolution**, then click **Run**. The clip appears on the node.

Workflow: A written scene and a Mood picker direct a Cinematic Avatar clip, which opens a longer edit.

- Text → Cinematic Avatar (prompt)
- Mood → Cinematic Avatar (prompt)
- Cinematic Avatar → Combine Videos

## Inputs
| Input | Accepts | What it does |
| --- | --- | --- |
| **Prompt** | Text nodes and pickers | The scene to render. You can also type it in the settings panel. |
| **Video ref** | Video nodes | Optional. A reference clip that guides the generation. |
| **Audio ref** | Audio nodes | Optional. A reference sound that guides the generation. |
| **Image ref** | Image nodes | Optional. A reference image that guides the generation. |

The output, **Video**, is the URL of the clip.

## Settings
| Setting | What it does |
| --- | --- |
| **Prompt** | The generative prompt, from 1 to 10,000 characters. It describes what happens; it is not spoken. Required. |
| **Avatar Looks** | 1 to 3 HeyGen avatar looks to feature. Required. |
| **Duration** | The clip length, from 4 to 15 seconds. The default is 10. |
| **Auto duration (let HeyGen decide)** | Lets HeyGen choose the length. **Duration** is ignored while it is on. Off by default. |
| **Aspect Ratio** | **16:9** (the default), **9:16** or **1:1**. |
| **Resolution** | **720p** (the default) or **1080p**. |
| **Enhance prompt** | Asks HeyGen to improve your prompt before it renders. Off by default. |
| **Pre & post text** | Text always added before and after the prompt. It is hidden from people who use your workflow as an app. See [Prompt pre and post text](https://nodaro.ai/docs/concepts/prompt-pre-post-text). |

Because the prompt is a real generative prompt, the Prompt Wizard and field mappings work on this node.

## Credits
The price depends on the resolution and the length of the clip. The editor shows the cost before you run.

- **A hold, then the real charge.** Credits are reserved when the run starts. The final charge follows the real length of the clip, and the rest is refunded.
- **Auto duration** does not know the length in advance, so it reserves the price of a 15-second clip at your resolution, and refunds the difference when the clip is ready.
- The cheapest clip is 4 seconds at 720p; the most expensive is 15 seconds at 1080p.

## When HeyGen is not set up

On a self-hosted install without a HeyGen key, the avatar picker is empty and shows a notice, and a run fails with the error `heygen_not_configured`. Add a HeyGen key or connect to Nodaro Cloud. See [Cloud connect](https://nodaro.ai/docs/self-hosting/cloud-connect).

## Tips
- **Direct, do not narrate.** Describe the scene, the mood, the camera and the action. The prompt is not read aloud.
- **One look for one subject.** Use 2 or 3 looks only when the scene needs several characters.
- **Draft short.** Start at 720p with a short duration, then raise the resolution and the length for the final clip.
- **Go vertical or square for social.** 9:16 suits TikTok, Instagram Reels and YouTube Shorts; 1:1 suits square posts.

## Frequently asked questions

### What is the difference between Cinematic Avatar and AI Avatar?

AI Avatar makes a talking head that says a script or a voiceover. Cinematic Avatar makes a short cinematic scene from a prompt, featuring 1 to 3 avatar looks, with no script and no voice.

### How long can a Cinematic Avatar clip be?

From 4 to 15 seconds, 10 seconds by default. Turn on Auto duration to let HeyGen choose the length.

### How many credits does Cinematic Avatar cost?

The price depends on the resolution and the length of the clip. Credits are reserved when the run starts, and the final charge follows the real length, with the rest refunded. With Auto duration, the reservation is made at the 15-second price.

### Can I use more than one avatar in a scene?

Yes. Pick up to 3 avatar looks when the scene has several characters. Pick one look for a clear single subject.