I have an automation that is triggered by a web form, giving users the ability to design their own yard signs.
The user inputs brand colors/text and I’m using OpenAI to generate the image.
My first OpenAI module generates the prompt and the second module generates the image.
This is all working fine.
But, what if the user wants to upload a logo?
How could I pass that to the prompt generator to include it in the design?
You mention that your users would upload data using a web form. This form will probably save this data somewhere, along with the logo file (eg. Google Drive).
Then, you’ll download the file and use it to generate the prompt. Depending on the model you choose, you get the option to pass an image file in the user message.
I haven’t tested this with OpenAI app yet, but I believe the model accepts this image because of its vision capabilities, meaning it will probably try to “read” the image. But it’s worth it giving it a try.
I do use this scenario a lot with Anthropic Claude, but only to extract info from the images.
I will try that out and see if it works. Thanks for the suggestion.
