fal integration: fast AI image and video generation for SaaS
Use fal.ai models in a ShipAny app — queue-based generation with webhooks, credit metering per task, automatic refunds on failure and results saved to your storage.
Last updated: Oct 8, 2026
fal hosts fast inference for popular image and video models. ShipAny's fal provider submits jobs to fal's queue and plugs them into the same AI task + credits pipeline as every other provider.
What ShipAny handles
- Queue-based generation through
queue.fal.run, with a webhook URL so fal reports back when the job finishes, plus status queries as a fallback. - Media types including image and video, with model-specific options passed through.
- Credits per task, deducted when the task is created and refunded automatically if it fails.
- Saving outputs to your configured storage.
Setup
- Create a fal account and generate an API key.
- In ShipAny, open Admin → Settings → AI → Fal and paste the key.
- Configure R2 storage for permanent results.
- Generate something in your app and confirm the task completes.
Tips
- Many templates route between providers — for example, the video templates can use fal or Replicate for the same model family — so you can switch when one is slow or more expensive.
- Price per model: set credit costs that reflect fal's per-request pricing for each model.