AI Facebook Ad Image Generation from Drive References

How it works

1. 1.Lists and downloads files from the inspiration folder and product folder in Google Drive.
2. 2.GPT-4o analyzes inspiration images for reusable static ad style principles and product images for core emotions and product identity.
3. 3.A third OpenAI node combines both analyses into a single prompt making the product the hero.
4. 4.An AI agent with a GPT-4 chat model and structured output parser produces ten JSON-formatted image prompts that Split Out separates.
5. 5.Loop Over Items routes each prompt through a 15-second Wait node then a POST to api.openai.com/v1/images/generations.
6. 6.Convert to File turns data[0].b64_json into image.png, Google Drive4 uploads it, and the loop returns for the next item.

Use cases

• Batch-generating 1:1 static Facebook/Instagram ad creatives featuring a specific product.
• Turning a swipe file of high-performing ads into reusable style templates applied to new products.

Requirements

• Google Drive OAuth2 credentials with read access to both source folders and write access to the output folder.
• A single OpenAiApi credential covering GPT-4o, GPT-4, and gpt-image-1 with sufficient credit for ten images per run.
• Replacing the three hardcoded Drive folder IDs with your own folder identifiers.

Setup and testing in n8n

1. Run manually and confirm both Drive listings and both GPT-4o analyses return content.
2. Verify the agent emits valid JSON with ten unique prompts before Split Out.
3. Check the target Drive folder for the ten generated PNG files and inspect the OpenAI API usage for errors.

Service Value

Ideal as a smart automation service combining integrations and AI to produce ready-to-use results.