Image generation not working

General
1

I’m trying to ultimately create an app that would allow users to upload a photo of their home’s landscape, then use prompts to generate different landscape variations for landscape design inspiration.

However, I can’t even get basic image generation to work. I have DALL·E 3 connected with my OpenAI API key and have it configured as a form. I keep getting the error “It seems there was an issue generating the image.”

I’m using a very simple prompt to test: Use the uploaded photo and improve by adding [user input]

Any ideas to what the issue is? I watched tutorial videos on YouTube and it looks like it should just work once you have your action connected and API key.

1 Like
2

Have you tried any of the other models like Flux etc,

I think these are so much better than Dall e and a lot cheaper.

3

So I am able to get the image generation to work with Flux. However, it isn’t generating images that match the original source image in the knowledge upload. I tried several prompts stating that I’m looking to improve the uploaded photo and that nothing else should be changed except for the specific landscape feature the user entered (ex. add green grass)

Is Flux not able to analyze photos and generate similar images?

4

If you send your tool over here, I can check it out and give you a better guess as to why the image generation is not working.

Without knowing specifics, a few guesses are:

  • You don’t have a proper OpenAI API key entered for the DALL-E action.
  • You are generating content that OpenAI is flagging as something it doesn’t want to generate.