Pet scene video

What does this workflow do?

2 image inputs pass through 2 sequential model steps to create a video. Inputs are uploaded as private assets; completed results are accessed through the authenticated API.

What should I upload?

• First pet photo (pet1): image, required, up to 8 MB. Accepted types: image/png,image/jpeg,image/webp.
• Second pet photo (pet2): image, required, up to 8 MB. Accepted types: image/png,image/jpeg,image/webp.

How are the inputs processed?

1. Edit the photos with fal-ai/nano-banana-2/edit. The editing step uses the configured input photos to create a single scene.
2. Generate the video with fal-ai/wan/v2.2-a14b/image-to-video. The video step animates the preceding image result.

Is there a verified example?

No example output has been published for this reference. The recipe diagram is a description of the pipeline, not a generated result.

This workflow produces video without an audio track.

What needs configuration before launch?

Enable product.trends.enabled in app/core/product.ts, review the workflow in app/core/trends/, and configure the supported models, verified cost rates and provider credentials. New generated products leave this module disabled.

Test uploaded inputs, successful output, failure handling and credit settlement before switching to a live product. Never retry a paid request whose result is unknown.

Follow the workflow setup guide · Read the API guide

Common questions