> ## Knowledge Base Index
> Fetch the complete knowledge base index at: https://help.seelab.ai/sitemap.xml
> Use this file to discover available pages before exploring further.
> Pure-Markdown content can be obtained by appending a '.md' suffix to the content URLs listed in the sitemap (without the trailing slash).

# The new Video space: from approved image to product film

Your visual is approved? Bring it to life. The new Video space brings together ten video engines, your Seelab models and a tool to edit an existing video, all in one place. A product turning on itself, an ambassador using it, a film that ends on your logo: it all starts from your images.

${youtube}[Video demo](cCIsjYRSaa0)

# Three tabs, three ways to create

* **Medias**: generate a video from a prompt and your images.
* **Custom models**: animate your product directly with your Seelab model.
* **Edit**: change an existing video without redoing it.

# Your product, intact from start to finish

Every trained Seelab model also exists as a video model: one training, two uses. The model guarantees shape, label and colors; you only describe the scene and the motion. For full control, start from an approved image with **Start / End Frames**.

# Up to 10 references in one video

With **Multi Refs**, bring together a character, your product and your end logo. Each image is numbered, and the prompt says what each one becomes: "the woman in Image 3 picks up the mist from Image 2… then fades to the logo in Image 1".

![](https://storage.crisp.chat/users/helpdesk/website/-/6/8/0/3/6803ea00be5f5800/galn3g18_6ndfvh.jpg =1000xauto)

# The right engine for every shot

| Your need | The engine |
| --- | --- |
| Animate your product without distorting it | Your Seelab video model |
| An approved visual, with sound | Gemini Omni Flash 1.1 |
| Fabrics, liquids, human motion, shots up to 30 s | Kling v4 |
| A multi-shot film with voice and references | SeeDance v2.5 |
| A product video up to 4K | SeeDance v2.0 |
| A transition between two images | Minimax H3-Max |

16:9, 9:16 and 1:1 formats, cinema 21:9 depending on the engine, up to 4 videos per run, and the cost always shown before you click **Generate**.

# Camera moves in one click

**Camera Presets** add a predefined camera move or style, without describing it in the prompt.

# Edit a video without redoing it

The **Edit** tab keeps the original motion and framing, and changes only what you ask: swap a product for the one in your reference, change a color, the setting, the time of day, the camera angle or the style. Import a 3 to 30 second video, or send one straight from your feed.

![](https://storage.crisp.chat/users/helpdesk/website/-/6/8/0/3/6803ea00be5f5800/galn3g42_18qst1c.jpg =705xauto)

*Left, the original video; right, the edited version.*

# A good video gets variations

**Reuse video setup** takes all the settings of a successful video to launch a variation. File your best videos in a collection to share them with your team.

**Your turn**: open **Video**, start from an approved image in your feed and describe the motion.

# Learn more

* [Generate a video](https://help.seelab.ai/en/article/how-to-generate-video-with-seelab-15wg3ir/)
* [Multi Refs](https://help.seelab.ai/en/article/multi-refs-character-product-and-logo-lc99ze/)
* [Animate your product with your video model](https://help.seelab.ai/en/article/animate-your-product-with-your-video-model-1f7ismq/)
* [Edit a video](https://help.seelab.ai/en/article/edit-a-video-with-example-prompts-7fi5gh/)
* [Which video model to choose?](https://help.seelab.ai/en/article/which-video-model-to-choose-kim2xb/)