Airtable image queue for design teams

Let your team drop prompts into an Airtable row and get a finished image back in a shared Google Drive folder, hands-free.

Deterministic Code
Airtablefal.aiGoogle DriveMarketingOperationsContent GenerationData Sync
PromptCreate

Turn an Airtable table into a batch image queue for a design or marketing team. This is a pure code workflow — every step is deterministic. Do not add reasoning steps.

Trigger: poll Airtable for new records in the Image Requests table (event type new_record). The table has columns for the prompt text, model choice, aspect ratio, style notes, status, error message, and the finished Drive URL.

For each new record, first check whether the Drive URL column is already filled — if so, skip the row so a re-poll never regenerates an image that already exists. Otherwise, read the prompt text and options (model, aspect ratio, style notes) off the row.

Call fal.ai Generate Image with FLUX.1 [dev], passing the prompt and options from the row. The result is a direct image URL on v3.fal.media that is reachable without extra auth.

Fetch the image bytes from the fal.ai URL and upload them to the shared Google Drive folder using Upload File (Multipart). Name the file after the Airtable record so it is easy to trace back. The 5MB multipart limit is comfortable for FLUX.1 [dev] output.

Finish by calling Airtable Update Record on the same row: PATCH the Drive file URL into the image URL column and set the status to Done. Only update those fields.

Error handling: if fal.ai returns a 422 (prompt validation failed), catch it, set the row status to Failed, and write the exact fal.ai error message into the error column so the requester can see why. Do not retry a 422. Let all other errors bubble up so the platform surfaces them.

Additional information

What does this prompt do?
  • Watches an Airtable table for new image requests and picks up each row as it arrives.
  • Generates the image with fal.ai using the prompt and options the requester typed into the row.
  • Saves the finished image to a shared Google Drive folder and writes the link back onto the same Airtable row.
  • Marks the row as Done when it succeeds, or Failed with the error message if the prompt was rejected.
What do I need to use this?
  • An Airtable base with an Image Requests table (prompt, model, aspect ratio, style notes, status, image URL columns).
  • A fal.ai account with credits for image generation.
  • A Google Drive folder where the finished images should land.
How can I customize it?
  • Swap FLUX.1 [dev] for FLUX1.1 [pro] when you need a higher-quality tier.
  • Point uploads at a different Drive folder per team, brand, or campaign.
  • Add extra Airtable columns (aspect ratio, negative prompt, seed) and pass them straight into the generator.

Frequently asked questions

Do I need to babysit the queue?
No. Once the workflow is running, it checks the Airtable table on a schedule, picks up any new rows, and processes them one by one. Your team just adds rows.
What happens if the same row is picked up twice?
Any row that already has a Google Drive link filled in is skipped, so a re-poll never regenerates an image you already have.
What if fal.ai rejects the prompt?
If the prompt fails validation, the row is marked Failed and the exact error message is written back to Airtable so the requester can see why and try again.
Can non-technical teammates use this?
Yes. That is the whole point. Anyone with access to the Airtable base can drop a new row and get an image back without touching the workflow.
Where do the finished images live?
Every generated image is uploaded to the shared Google Drive folder you connect, and the direct link is written onto the Airtable row.

Turn Airtable into a self-serve image queue.

Give your design and marketing team a shared table where prompts go in and finished images come out, automatically.