Fal.ai video Generator
Fal.ai video Generator creates short videos with the video models offered by Fal.ai. You can start from a written description or from an image you want to animate, and choose among several models depending on the style and format you need. The component waits until the video is ready and returns it so you can keep working with it in the flow.
Use it when you want to try different video models from a single component, or to produce clips in series for marketing and social media.
When to use it
- Create a short video from the copy of an ad or a post.
- Animate the image of a product or an advertising poster.
- Compare the result of several video models with the same description.
- Generate one video for each item of a list, for example one per weekly offer.
- Give an agent the ability to create videos when a user asks for one.
Configuration
This component needs a credential of the Fal.ai API service. See Credentials to create one.
| Field | What it does |
|---|---|
| Credential (required) | The credential of the Fal.ai API service. Pick one with Choose an option or create one with Add new credential. |
| Generation Mode | How the video is created: text_to_video (from a description) or image_to_video (from an image). Default: text_to_video. When you change it, the Video Model list shows the models for that mode. |
| Video Model | The Fal.ai model that creates the video. Default: veo3/fast. In image_to_video mode, pick one of the models the list shows for that mode. |
| Aspect Ratio | Video format: 16:9, 9:16, 1:1, 4:3 or 3:4. Default: 16:9. If the model does not support the chosen format, one it supports is used. |
| Duration (seconds) | Video length in seconds. Default: 5. Each model accepts its own lengths; if it does not accept the one you set, one it accepts is used. |
| Prompt | The description of the video you want. The maximum length depends on the model. You can type it or connect a text from another component. |
| Image URL | Appears in image_to_video mode. The web address (URL) of the image you want to animate. |
| Image Files | Connection point for images coming from other components. Models that work with several images use them as reference. |
| Negative Prompt | What you do not want to appear in the video. Only some models take it into account. |
Batch processing
With the Batch Mode switch on, the component runs once for each item of the list you connect to Batch Input and returns all the results together as a list. Parallel Executions sets how many items are processed at the same time (1 to 5). Skip if Empty Batch Input skips the component when the list arrives empty, and Skip Branch if Empty Output stops the following components when the result is empty. More in Key concepts.
Using it as an agent tool
You can connect this component to an agent’s Tools input. In the Tool tab, Tool Name and Tool Description are the name and explanation the agent uses to recognise the tool and decide when to call it: write a clear description of what it does.
Output
- Video: the generated video in MP4 format, with its file name. Pass it to the next components of the flow to save it, send it or keep processing it.
- Tool: the component as a tool, to connect to an agent’s Tools input. When the agent uses it, the video is saved with the conversation and the agent receives its name and the link to watch it.
Example
The supermarket chain Mercado Alameda publishes a short video with the star offer every Monday:
- Add Fal.ai video Generator and choose the Fal.ai API credential.
- Leave Generation Mode on
text_to_videoand theveo3/fastmodel in Video Model. - Choose
9:16in Aspect Ratio to publish it in portrait format. - In Prompt, write: “A basket of fresh seasonal fruit on a wooden counter, morning light, the camera slowly moves closer”.
- Run the flow and connect Video to the component that saves or publishes the file.
Good practices
- Try the same description with two or three models and keep the one that best fits your brand.
- Describe the scene, the camera movement and the light in concrete sentences.
- Try a single video before turning on Batch Mode with a long list: each video uses your Fal.ai account.