Feature Requests

Implment a way to help the AI know what you want. For example, I was trying to replace a chair I was sitting in with the iron throne, throne of swords, swords. It try but really had no idea.

Hi I think it would be a great idea to introduce a daily / weekly contest where users can submit their creations. You could announce a new subject each day / week (eg. Dragon, Cyborg, etc.) and allow everyone to submit entries. Winners and runner-ups could be awarded with badges. Not only would this keep people coming back, it is also great for providing users with inspiration and they play around with different tools in an attempt to get their image looking perfect, learning more about Firefly in the process. Thanks

Add Detail, Boldness & Extrusion sliders to allow the artist to adjust their final text appearance to meet the requirements of the reader/viewer.  (Some endusers or artists may want a more sedate or more extreme final product view.  ie more or less fancy or extreme look)….just thinking….  

I don't understand why people are not talking about procedural material creation using Generative AI! It could open gates of infinite possibilities, what if we can create AI variation from nodes output or we can previsualise before even getting started in desinger. What do you guys think? 

Adobe Firefly is an AI tool that can generate illustrations from text prompts. With Adobe Firefly, you can create custom illustrations for your projects, generate concept art for your games or movies, design new products or marketing materials, or simply have fun and explore your creativity.To use Adobe Firefly, simply type a text prompt into the interface. The prompt can be anything from a simple description to a more complex idea. For example, you could type "a painting of a cat in a field of flowers" or "a futuristic cityscape." Adobe Firefly will then generate an illustration based on the prompt.Adobe Firefly is still under development, but it has already generated some impressive illustrations. The tool has been used by artists, designers, and creative professionals to create everything from concept art to marketing materials. Adobe Firefly has also been used by individuals to create personal projects, such as creating custom illustrations for their children's books.Here are some

after an Image has been generated in Text to Image, an Icon can show up, enabling the user to edit the generated image in Generative fill. 

life is going very fastand AI is Very Fast technology too , but in all AI i seen just text to image i think its the same technology with deffernt name , but if adobe can convert voice to image , it be like bomb .......like an experience in business management and design i think its easy step but also its very Hug step for adobeMakkin Ibrahim

streamline tasks and improve the outcome of my work

I would really love a feature where we can specify the part of the AI-generated art that we want to change. Instead of having the AI generate new pictures each time, with completely random changes, I want to be able to specify which parts I have a problem with. For example, if the hands are not realistic, I want to be able to select them and generate images with just the hands changed (everything else the same). Or, if I really like the picture, but want to the see it in a purple hue, i want to be able to do that. I understand it might be really difficult to implement though. 

We need PS Beta with AI in the IPad Pro and the full Premier PLEASE naw is the time !

Hii, i have been using the Text to image and the generative fill in sync and have generated a lot of images. Throughout the process i suffrered from ine thing which is perspective. When i try to replace of modify somthing in the foreground or background. It does the filling and that too it is good.Although it struggles with the perspective and the scaling of the image.As seen in the image i attached 

A beautiful girl with closing eyes and having all natural things on her hairs. Lying on a river l.

the ability to fit a in a signiture that is part of the art or scene itself both hidden signutes and and bold ones for marketing 

My idea is that Generative Fill is provided with Text to Image style options to get the ideal background.

Hi there, Great addition would be to have an option to remove a background and specify to remove a spill of color or how hard should cutout be. Many Thanks

hello, it would be nice if adobe firefly could remove flickering from video or even had a video quality enhancer tool

Why image generator can't recognise or generate text correctly?Maybe some image is fullfiled to idea that you imagine, but only one problem: text placed on that image is terrify...Example in attach

After the beta period is over, firefly will be open for commercial use. We may use it for generating web design, app logos etc. For such tasks we might need the best result out of all the generated images. Since diffusion is a random process, once we click the refresh button the earlier image is lost. It's impractical to download a dozen images and later select best or open 10-12 tabs simultaneously. So I'm suggesting a compare feature similar to those found in shopping sites where we could place 5-10 images, pick out the best image and later remove the remaining images. This woud save a lot of time and could potentially be very useful.

When using the "text to image" functionality, images often contain text and the text often contains spelling errors. Having the option to omit the text could be really beneficial, e.g. a checkbox or specifying a promt like "Create an image representing X where no text is shown on the image".

Hello,I've been an architectural draftsman for several years now, and your Adobe tools have played an integral role in shaping my career and bringing my creative visions to life. I find myself constantly inspired, and this inspiration has sparked some ideas that I believe could enhance our shared creative journeys.Imagine if Adobe Firefly could help us not only envision, but also sculpt the skylines of tomorrow. An additional style focused on realistic architectural forms - clean lines, distinct structures - could make our dreams tangible, right there on our screens. The impact on our work would be revolutionary - less time spent on tedious processes, more time igniting our imaginations.On a personal note, web design is a passion I nurture in my spare time. I believe many of us here share this interest, and we're always looking for ways to make our creative processes smoother, more intuitive. Imagine a world where Adobe offers a unique style tailored specifically for web applications,

I created really good textures with the AI and then was trying to get the AI to apply it to a suit. It would be interesting to have options on how the Reference image is used with your prompts; such as texture, background, subject, ect... 

le personnage dans l'affiche doit etre le youtubeur francais yu gi oh volht 

Si agregaran algo relacionado con el anime, podrían abarcar mucho más mercado, sin problema alguno, además de variar y expander exponencialmente el contenido que ofrecen.

Sad boy walking in rain and listening music

need to improve on traditional clothing images