How do I talk to Firefly?

Hello @aliBrad,

I will try to answer some of these...

How can I tell Firefly to explicitly NOT do something? 

The model does not understand the concept of a word being negated by the word in front of it. So using "not [X]", "no [X]", "exclude [X]", "with no [X]", and "without [X]" does not work. It ignores the negation word and adds to the image the thing you are trying to avoid.  You basically need to find "positive" words and phrases telling the model what you want to see rather than telling it what you do not want to see.  Such as "flat land" rather than "no mountains" or "desolate" rather than "no people".  If I am not sure how to describe something I want to exclude in a positive way, I will ask Google Gemini or ChatGPT for ideas.  Just be aware that when they give prompts, they are usually too wordy and conversational, so you have to edit them, but they are good for ideas.

How can I distinguish between a description of what I want to see and the description how I want it to be?

This is a hard question to answer and we could probably help better if you gave some examples.  My experience is the model responds best when you can describe what something looks like, rather than its state of being. One notable exception is emotions in people and animals.

When will there be simple and basic functionalities like "zoom out/ zoom in" "transparent PNG". 

These have been frequent requests. Adobe generally does not say in advance when something is coming–it will be announced when it is here. I recommend you upvote these posts to show your support for these features:

• P: Transparent background or layers 
• Zoom in on images / Higher resolution 

When will it finally be useful for something and not just produce random stuff?!

While Firefly does sometimes seem to have periods where it appears to be struggling, I use it every day to produce images I use in my work. Learning the best techniques for wording prompts makes a huge difference. Also, when you cannot get exactly where you want, taking the image into generative fill or Photoshop can get you to the finish line. What helped me learn a lot last year was attending the daily (Monday-Thursday) Prompt Jockey sessions hosted by Shawn Kosel and Jack Watson on the Firefly Discord. They work through problems and demonstrate what works and what does not work across the tools with Gen AI capabilities (Firefly web, Photoshop, and Illustrator).

Good luck!

My best,

    droopy