#community-chat
1 messages ยท Page 9 of 1
Everybody in mode FOMO in 4,3,2.....
thanks - I missed that in the faq
A lot of people have applied, see the latest information in #announcements , as long as you have submitted the form, are over 18 and have used your Adobe ID to sign up, patience you must have. @weak wren
if it works then ik what I did wrong in my previous attempts
Adobe folks, for those of us in and creating art, what do you want from us? Is our primary function as testers to identify where the AI is having issues with prompts aka the thumbs down?
I am sure they appreciate feedback
I understand the popularity of Firefly, my area is now 11 a.m., and I still stick to the website to keep refreshing๐ญ
lol
They have a section for features an ideas on here. You can submit them if you have one.
I actually couldn't stay in bed till 10am like usually to much excited
Submitted my first one a few minutes ago!
Probably just wait for the email or you might wear out your F5 key. :D
yes, I think it is important to give feedback not only on the good results but more importantly the bad results so it can learn from this
During this Beta, feedback is the most important part, yes. The ๐๐ feature is an easy way to help the team understand where there are still issues. But for us it is equally important to just chat with you all here. Let us know what you like and what could be done better.
The F5 will learn what i means to be as abused as some wasd keys are.
I'm impatient!
hands and limbs could be done better
is there a feedback tool for the AI for personal favors? so the AI learns what you prefer? or nah
probabaly not since everything is in the cloud
I don't know if the thumbs up is for that or not
maybe, but I am talking more advanced
could suggest it as a future feature
yep, but how to phrase it properly tho
I had to go to bed and hopefully tomorrow morning I would be able to use the fireflies, good night from China, although maybe everyone just got up!
it's afternoon for me
Be prepared for a million suggestions I'm almost positive you guys are already well underway on behind the scenes.
listening ๐
I'll be on the forum!
I've been up since midnight here so around 15 hours so far, most of that entering prompts
just finish running already(talking to this stupid google collab cel)
Unbelievable! Work better with a good rest!๐คจ
Mainly I have tried most of the AI painting tools so far, but they are not very satisfactory to me and do not help me very well. But some of the Firefly advertised were very appealing to me, so I was eager to experience them!
Good morning, friends!
see, the bot works ๐ฎ
good morning !
Woop woop, we got a bot that is alive!
It's already midnight in my area, and I'm still refreshing my website to see if I get Firefly permissions!๐ญ
Hey @slim lion - welcome back!
patience @weak wren ๐
It's been about 48 hours since I submitted my application! Can you help me๐คฏ
It's over 48 hours for me as well, it happens when it happens
gm Howard
Okay, I'm going to be patient!๐ฉ
You're certainly not alone on the wait list - we're approving people but that takes time ๐
I'm starting to lose confidence that I've requested access. Can I apply again? Will it slow down the process if I apply again?
I think it just needed its elastic bands wound up again
The "Use as reference image" feature is pretty dope, <@&1076199634810196038> :)
Does the approval of authority have anything to do with the region? I'm from China, and none of my Chinese friends around me have permission approval from Firefly
I think it will
I don't have any information on that, unfortunately.
You certainly don't have to apply again, this will not speed up the process but probably will make it even slower for everyone. If you have submitted the form, you are on the wait list ๐
I'm excited to begin and see what prompting is like here and how it differs from Midjourney. The starting philosophy here seems really interesting! Waiting will be worth it!
yes! But a bit hidden, I think
Very
@slim lion could use a tip about this perhaps!
That was the first thing I covered in #tips-and-tricks. It's a great feature, just a bit hidden for now.
It's not "obvious" but I wouldn't say its hidden. :)
Can't wait for the ability to upload and inpaint my own image. :)
Because you can see on the channel #archived-share-your-work-3-23-to-11-23 , almost no Chinese designers get the permission to firefly, which has to be a pity
just because you don't see posts from chinese people doesn't mean there aren't any users. They might just not be posting their work on social media ๐
I see what you mean! I don't mean anything extra, probably because none of my friends around me have access to it! In fact, everyone is eager to experience it!
Good Morning @slim lion
I just requested access, this is so good from Adobe's part.
hey all ๐
Welcome!
What are we creating today?
It will be so nice to take an existing headshot/portrait, put it through a AI generator and give clients amazing and artistic stuff. Also, what's to stop a client from taking our photos and running it through some AI?
AI is distruptive! - LOL
How to get access to firefly
looking at #archived-share-your-work-3-23-to-11-23 , the question should rather be "what aren't we creating today"
lol
I'm definitely in the OMG ADD EVERYTHING phase of my creative journey - trying to figure out aesthetic color and lighting all at once is daunting
How much time to get access?
Given the high level of interest, we're not able to get everyone access as fast as anticipated. Weโre admitting on a rolling basis and as quickly as we can
I think part of the problem is that users donโt get any confirmation email when they fill out the form. I applied twice because I was afraid I misspelled my email address the first time. Sorry ๐. I was expecting to get a confirmation email as usual and when I did not get it I got a bit anxious. Everyone is way too excited to try this and I guess itโs hard to be patient ๐ฅฒ
Hi
Welcome ๐๐ป
Can a clothing company - Gucci, HM, Zara, Gap, Old Navy, etc - simply use AI to create their ads and forget all about us creatives?
It's just another (very powerful) tool in the belt.
If you're creative, you'll find a creative way of using the tool that the execs wouldn't.
We keep telling ourselves that but honestly if I needed an ad today (or in the future) I'll do it myself with AI, LOL
Yeah, but AI is still utilizing your creativity.
True.
Painting didn't die because the camera was invented.
It definitely changed dramatically, but it didn't die.
how to get access to FireFly faster? โ I wanna review this on my YT
I already have an adobe account just not on the email I signed up for firefly with
I applied like 2 days ago
Could they? Probably. Will they? I honestly don't think many will. There's something special about that a human can provide that AI simply cannot yet. I think we're going to see campaigns that contain a lot more 'meaning' behind them to separate from the AI--created content.
Hello all
The key word here is โyetโ, I think! It will become better and better. I donโt think it will be able to completely replace humans though. No matter if we want it or not, itโs here to stay, and you can either fight it, or use it to your advantage, as an aid to speed up your process and give you ideas. With Adobe entering this market, I hope that at least initial artists would be respected.
Thatโs great ๐
When I heard about firefly I was so happy . Adobe really is the best ๐๐ฝ
Hello everybody
@slim lion - is there any switch or syntax to do "negative prompting"? For example, --no faces, people, etc?
cant wait to get firefly access
hello
Oh I'd love to know this as well
That's my view, as well. The AI turtle has quickly turned into a bullet train. We now need to figure out how to use it to our advantage, and that's what the various tools we're building will do.
Currently we use: [avoid=green, cars]. This is subject to change, though.
Yeah, this is how I envision using it especially the Adobe's Firefly since we already have Adobe for Photoshop, Premiere, After Effects, etc. If I can take my client's work and use AI to make quicker edits, or put my model/subject in another environment, etc that will be amazing.
i need the permission to use this
i'm on the waitlist, be excited to start trying it out
have you submitted an application on https://firefly.adobe.com ?
Very excited to try this out
I think at the tail end of the AIs development creatives will still be needed in a different capacity to where they are now. You'll probably see things in the graphical professional space being something like "prompt engineer". Give it ten years, and we'll have a better scope of how AI is going to affect professional creatives. But as stated before, you can either embrace it and make it work for you or continue chugging along in your current sphere.
I love that the model understands context, "red shoe, green ball, blue box" and gets it right a majority of the time. This is really important. :)
After we have submitted an application for the Beta how long should it take to get the email?
human-generated art will still be useful for training AI for a long time. the amount of art that has been generated relative to the information it expresses is very small compared to the corpus of text, and there are lots of reasons for that
Hi @fallow edge : See #announcements where an update was posted today.
Since there currently is a high level of interest, we're not able to get everyone access as fast as anticipated. Weโre admitting on a rolling basis and as quickly as we can ๐
a man in gold rush
PSA! If someone helps you get an answer, and you see the same question somewhere else, please pay it forward
โค๏ธ community is all about helping one another โค๏ธ
generating content is not supported in Discord. please check the #faq for instructions on how to get started ๐
You should first give access to all the YouTubers and Bloggers with significant audience to create a buzz
not to random people
doesn't make sense to generate buzz if they can't handle the influx
Are you a "Youtuber, glogger with significant audience"? - LMAO!
blogger*
It does make sense because it will generate a whole lot new customers for Adobe
That just complicates things. Then we'll have anyone who has a YouTube channel applying and clogging up the requests that way.
There's already more than enough buzz, judging by the number of people trying to join
Currently Adobe is losing a lotta of them to Canva and tools like that who already implemented some AI tools
Hi everyone! So excited to see some of the work from you all ๐
โLet me give you a hand!โ
Exactly. I believe it's best to allow paying creatives access first since we already use Adobe's suite of software for our work.
#/image beautiful rainbow paw cat pattern
You cannot generate in discord, please see #faq and take steps toward getting the access to Firefly Beta
We're focused on a smooth beta experience and the team is working as hard as they can to get there.
"Canva is a more user-friendly and affordable option for beginners and small businesses, while Adobe Express is a more powerful and professional tool for advanced designers and larger organizations."
translation: we get that you're a "big important influencer" but you still have to wait your turn like everyone else
I don't think Adobe and Canva can be compared, they're in different games (yet similar).
True but you could prioritize them somehow, I'm your paid customer, and I'm also your affiliateโyet I haven't gotten access to Adobe FireFly yet... I think it's quite unfair
Im way too amped about firefly....
Hello, I registered to have access to the beta, do you know how long it takes and how the selection is made?
Can one ever be TOO amped!? ๐ค
It's first come first served but we're backlogged and some applications are taking longer than others. So it's tough to say. Hopefully soon!
id say! When you cant finish explaining ideas cause your brain is exploding with more and better idaeas.... Thats toooooooo amped...
even my grammar is affected ๐
ok thank you and also I registered with my personal email address and not the one I use for the adobe suite which is my work email does this affect acceptance
Combine excitement for the possibilities of new tech and a few cans of Red Bull... it's possible. :)
the interface is inspired, i enjoy being able to alter the preprompt with the same seed
Speaking of the Firefly UI, what else would you all like to see?
there's no kind of weighting or multiprompting yet, right?
sorry forgot to turn pings off
Nope, but that just gave me an interesting idea...
i'm trying to prompt a character with just-above-the-shoulder hair and "short" is too short while "medium" is too long, and ordinarily i would weight down short, or multiprompt one medium and one short
i am really exciting to finally use this software, i think it is easier than other ai softwares out there, just perfect for me.
i wanna know that too
hmmm, "shot from above" and "shot from below" don't seem to have much effect on portraits (probably not a lot of portraits in the dataset that aren't dead level)
If you've applied on the site then you're in the queue. But we're really backlogged so it might take a while. And you have the wrong discord for Midjourney.
This server got 37.000+ members in a matter of few days
Excited for this. Hopefully will gearing up soon
It doesn't affect acceptance but the invite will go to the email you entered in the form.
i submitted my email yesterday morning and got my invite early this morning, so its not a huge wait
Seriously? I submitted mine Tuesday afternoon...I've been waiting nearly 48 hours. First come, first served though...
ok thanks!
Hopefully soon! The server has 37k members if that gives you any indication as the number of people applying. And unfortunately it's a pretty manual process. I don't think we've anticipated this kind of hype.
I get it, but you keep telling everyone one thing and it's clearly not true.
yeah, I don't want to know how many people have applied, who didn't even bother signing up on Discord ๐ฑ
getting a lot of these when prompting someone "wearing glasses" 
I was excited...getting kinda frustrated. Not because I have to wait, but because some of us are getting leapfrogged and nobody wants to admit it.
40/40 vision
its so they can do the dramatic taking off of glasses and still see
I would put paying customers first tho. ๐
I would as well.
I've been using Photoshop since 1990, on a 68k Mac. Where does that put me?
And you too have no access yet?
I SO LOVE THE TEXT EFFECTS IN ADOBE FIREFLY
Not yet.
Me too! Photoshop 2 lol
Please show us your receipt from your Photoshop 1.0 purchase ๐
i don't think they are in a position to verify pre-creative-cloud claims
The days when modem speeds were measured in k
imagine how much longer it would take if the team also had to verify if someone actually has been using Photoshop since Ps 2 ๐ฎ
If only I still had it saved after all these years!
Right?! That's something I would frame
PS 6 or 7... i had to buy a paper book to learn how to use it...
I think my account still has my CS4 and CS5 license information contained within. :D
Firefly was the best show tho.
Yes!! I was so heartbroken when I found out Martin Evening passed. I learned a ton from his books.
Anyone get their invite yet? Join first day, just patiently waiting!
"Please submit your receipt for Photoshop 1.0, your shoe size, favourite pizza topping and the name of the character you never liked in that one kids show you always used to watch"
hoo boy do not prompt a woman with freckles in photo mode until weighting is a thing, trust me on this ๐
I started using photoshop more seriously in school around CS5. But my dad had photoshop 1.0 that I used sometimes and he used to make invitation cards to my birthdays in an early version of illustrator ๐
"Pineapple on pizza" fans get priority access.
btw is the invite FIFO, or should apply with all of our company's accounts to have better chance? ๐
that was supposed to be a secret
@quasi valve I think it may be a while.
I had that three days ago lol
I used Photoshop in Windows for Workgroups as a wee lad. :)
pineapple pizza fans
@terse raptor AHA! the plot thickens
I'm going to use this for the example I'm mocking up. Is the idea here to reduce the 'intensity' of the freckles?
now I want pizza ๐
yeah
Is this a trick to kick all fans of pineapple on pizza of the list? ๐ it should I add something outrageous like โI like curry, pineapple, banana and chicken on my pizzaโ
novelAI (and i think most other stablediffusion forks) have the same issue, which you handle by weakening freckles
You guys use other AI Generation platforms? Midjourney and SD and my favorite, but really looking forward to this!
but even prompting "light freckles" gets you, uh, this
I'll post a (very unofficial) UI design idea here at some point today
Howdy Doody's daughter
Holy cow!
Peppy Long Stocking didn't age well? LOL
i don't know if text-shorthand is against the design principles or whatever, but you might consider adopting the {strengthen} and [weaken] syntax that most stablediffusion forks have, just because its so universal (and because weighting individual words up twice, down once, up thrice can probably get pretty messy in a UI, as opposed to just nesting them for words)
Was that actually generated with Firefly though?
It doesn't have the watermark.
((if you can make that kind of micromanagement look sensible i'll be impressed))
it was i just did a screen capture instead of a download/upload because it was faster
It's tempting to go down that route, but I'm going to see if I can come up with a UI solution vs text, as a fun challenge ๐
its asking me to request access
it does sound fun! but really hard
i feel like the ai needs negative prompt 
oh yeah, negative prompts are a big thing too
assuming this is a clip-guided diffuser of some stripe
i am not able you use any of the ai tools on firefly
what does request access mean.. is it available for people or not?
if its a GAN, well, i don't know jack about GANs 
You need to request access to the beta and you'll receive an email once approved.
Check out #faq for more info
oh okay... thanks buddy
is this enough pineapple
No.

here you go, the ideal adobe pizza

more of a calzone really but whos counting
- maintains eye contact while picking the pineapple off the pizza *
I'd eat this
did it just add those candles spontaneously
that is literally a pineapple pizza. haha
The perfect romantic night
"this dinner sounds romantic, better amp it up"
I feel like someone needs an intervention
are the "graphic" and "art" settings based on some internal metadata, or is it different preprompts? because they seem quite similar with the prompts i'm running so far
No banana? Iโm out!
but i've been using this for *checks watch* 15 minutes so i've probably missed something
i assume "graphic" would be going for clean lines and illustrator-type stuff, and art more of a noisy brush situation
They're similar, but the 'art' styles go well with many of the 'techniques', like Oil Paint, Doodle, etc.
Fiiiiine.
Chefโs kiss ๐๐ป
so atm people who start their name with Lo and love Pineapple on pizza get faster acces,sweet getting closer and closer by the minute
lol
I'm both intrigued and disgusted at the same time. Its not appetizing and yet, I can't look away.
swap
I'd eat
what's the prompt?
another issue that i definitely know is tough for these kinds of high-powered image diffusers to solve: humans with non-human skin (for SFF settings and the like)
Editorial style photo of a pineapple with pizza and a banana on top, on a rustic wooden table, rustic tavern atmosphere, dimly lit restaurant environment, restaurant in the background, premium stock photography, highly detailed, rustic style
(weighting might help with this too, when it's in)
so to get access one needs to sign up for the beta? seems like ya'll got access quick
Yep. Waiting waiting waiting here...
the more coherent you train it to be with humans the harder it is to get it to break those rules in a prompt
so that's what the banana tool in Photoshop is for
or do i have pay to acess firefly
Nope! Just need to request access. The team is working through the requests and you'll receive an email when you're good to go
So I should return my Chipotle gift card?
sweet. thats nice cause i have no real use for it . other than just to play with it. lmao
alright i have to stop playing with this now or ima lose my day job
xD
Same. Oh wait...
its very good for a first beta, most of the problems i've noticed are pretty universal struggles in midjourney/stablediffusion/dalle
I told my boss I would only spend 1/2 hr playing with this, good job i'm the boss
Did vector text using Photoshop before Adobe Photoshop had vector text. But I have access. Seems to be at roughly midjourney version 2/3 or so in quality right now.
i'm just a dumb admin lawyer who plays with AI art in their free time, i can in no way justify this for work
Isn't this your day job? Xd
๐
I've got plenty of time to play around with this stuff
So I sent my request to try out the app. Did you all get an email saying the request was received?
Try to generate some lasagna
@slim lion what did you study to get here btw? Xd
what
Some did, some didn't, but as long as you submitted, you're on the list!
Went to school for 3D design/animation. That didn't work out. ๐
Euhmm.....
I am doing AR Editing and User Experience rn
i wnt to school for network system administration .. and you know whati do ...clean boats
Part of the webdesign degree
ai
@slim lion you know any tools that transfer 2D characters (avatars) to 2D meshes? That I can easily use without errors
@slim lion ไฝ ๆฏ่ฐ
I don't ๐ฆ
here's another example of a thing that gets weirdly hard for AIs to do the more coherent their data is

What do you do while you are waiting for your access to Adobe Firefly?
Here are 3 Feature Ideas for Adobe Firefly that will help you during this hard timeโuntil you get accessโkeeping you entertained ๐คน
Text-to-Smell: With this feature, you can input a text description of a scent, and Adobe Firefly will use AI to generate the corresponding aroma. Imagine typing "freshly baked chocolate chip cookies" and being able to smell it through your computer!
Text-to-Taste: This feature takes Text-to-Smell one step further by allowing you to input a text description of a food item and generating the corresponding taste in your mouth. Type "spicy buffalo wings" and experience the heat on your tongue without even having to leave your desk!
Text-to-Teleportation: This feature may sound like something out of science fiction, but with Adobe Firefly, it could be possible. Input a text description of a location, and the AI will use advanced algorithms to teleport you to that exact spot. Want to visit the Eiffel Tower without the hassle of travel? Just type it in and let Adobe Firefly handle the rest.
damn you got access already? Is MJ bot still down?
CHANGE BLUE SKY TO RED
yeah i applied yesterday morning and i dunno, but don't bother non-MJ discords about the status of the MJ bot it gets bad enough in there 
lol
hah coming soon, love it.
try some halftones with firefly, I wanna see how bad they are
@true bone you cannot generate images writing prompts in the discord, please read the #faq to learn more about firefly
Heck yeah guys i just finished submitting my request form out. wish me luck lmao
let me check their presets before i prompt it
hmmm "pop art" might be close
It's always hard to figure out if theres different words or styles that will evoke it, as I think most training on "halftones" isn't really it
well they have a list of, essentially, photoshop filters, so if anyone can do it they can (halftone isn't on there though)
How long did it take most of you all to get accepted ? jw lmao
yeah like that, comic book, or how people try dot-matrix etc. I've never found some that really work best, as its such a vague term that ends up mostly with bad examples in any training data, or more of the artsy style rather than some precise stuff
okay i ran this prompt with "art" and "pop art" on and yeah its not quite there yet
The resolution of the training and having good training images, as well as the resolution limit of the outputs, is what really holds it back...
But yeah thats actually not too bad its sort of getting the idea a bit
i don't think adobe has that problem considering their library
Well no I mean anywhere in the world, there are actually not a lot of examples of really high-res true-halftone conversions of images and artwork.
ohhhh yeah
Those examples are the "artsy" kind
But also the output resolution.... if it needs to be like 12,000 pixels to really capture it,, its gonna be a while before we have those kind of outputs
here's what happens if i take off the "pop art" preset
interesting, yeah popart was making it more smoother and those colors
you can see the sort of thing MJ was getting in the training and others
they end up seeing the low-res patterns as cross-hatches
if i take off the "art" preset entirely i get a bad-looking photobash with only background dots
but thats still pretty cool and good progress towards it, eventually we'll probably be able to custom-train these on our own high-res images for high-res outputs
let me try "graphic" just to see what it does
woodcut usually gives the other style of line-art instead of halftones, sometimes including vector halftones makes it cleaner but shifts more artsy
yeah nah art's the way to go here
its kind of like expecting a machine that pumps out round wheels to be capable of generating a combustion engine, but we'll get there eventually
Hello everyone! I just wanted to ask how much time did it take to get access to use Firefly? I've requested access a couple of days ago and I haven't got the email yet.
Guys, Registered with Adobe Firefly (Beta), not working why?
how do i get a job in adobe? I'm on my second year of graphic design
please
You've registered in Adobe, not in the Beta.
now that you have your adobe account, you can request access here @wooden granite https://firefly.adobe.com/
how much time does it normally take to get access ?
howdy, just signed up for access
same now we wait
we can't give estimate about this because the team still hardwork for this
Looks like you registered the Adobe ID. That is the Key to all other Products. NOW you can go over to https://firefly.adobe.com/ and register for the beta program. have fun.
no idea how long we wait
1 , 2 or maybe 3
lmao
@shut dew Here's a very quick and rough mockup of what I'm thinking for adjusting weights with the UI.
Disclaimer: This is very unofficial and I'm not directly part of the Firefly design team.
Beta applications are being approved on a rolling basis and you'll receive a confirmation email once you've been granted access. At this time we do not have an accurate estimate for when access will be granted. Thank you for your patience! #faq
Right
guys, please, Where will I click
that's clever but a big oof for the colorblind
Obviously this would need to go through accessibility checks ๐
?
First you have to "Request access" and after it is granted you can "generate" your work. Have fun.
also it might get really annoying to have to do it four times if you wanted to weight a four-word phrase
(i'm assuming you could highlight stuff and do it that way this is just a quick and dirty exhibit)
A man with curly hair
I would imagine you can drag over the words you want connected
you would just select the whole phrase and choose the weight
yeah its cool
accessible version might be something like thin/thick underline for modest/big overweight, dashed/dotted underline for modest/big underweight?
THAT Feature I wish also for the Text-Effects to get bigger or smaller Details of the Effect onto the Surface. Nice One.
i was watching the C*nva create event, and I saw many thing that can be added to adobe express if adobe manage to connect it with Firefly somehow.
Who has a channel? Give me a channelโน๏ธ
your in a channel
Hii! when can we get an access to adobe firefly?
Hello from Temecula CA
Express will definitely see Firefly integrations.
Hi there-how long did it take for you all to get access after you registered?
For those asking about access, the team is working as quickly as they can to get everyone in, but it's taking longer than expected. Hang right!
zero idea i did it about 10mins ago
How to generate pictures๏ผ
Off to lunch, catch you all later!
later
You need to join the waitlist and wait to get acces to the web app
Great-thank you. Looking forward to it!
oh is it a web app
taht i didnt know
i was figureeing i was going to have to download a huge program
i'm hella impressed that i gave an example of a useful feature request and there's a mockup of it one hour later
thats hustle
That means I have to wait for the official mail, right?
I already registered from here. I received a letter in the mail. But how do I start?
yes
that's just you registering an adobe ID, gotta wait longer to get access to this specifically
hope it's that pineapple with a pizza topping
ok ่ฐข่ฐข
it took them about 24 hours to get to me, but that queue might be longer by now
dang
Thank you
i just filled it out 10mins ago
ok ; thank you Shambibble
yeah its definitely not dominos pizza times
Do you have a video how to register?
more like amazon prime times
I signed up shortly after release was announced and still waiting so I assume if you have just recently signed up it will be a bit longer as some of us who requested access day of are still waiting
f
The queues are random. They're telling people it's first come first served but it's not. I've been waiting nearly 2 full days, and some people signed up after me and are already in. They won't acknowledge it, but that's the reality.
That is why I went ahead and did my personal account and my enterprise account just in case it was random
@wooden granite visit https://firefly.adobe.com/ to request access
and make sure you're logged into that adobe account you created when you do
I signed up one hour after the email announcement. It's like getting to the restaurant early, and then being told you have to wait three hours because a bunch of people showed up after you.
Flyer design for a delivery company, requests for a parcel man, location map, high resolution
Well who knows, you might get lucky.
But don't expect a straight answer lol
When will we get commercial rights?
Probably after beta
after beta , im sure it wont be free and beta users will have to pay ?
How long the Beta would last?
Can't do prompts in here, my man.
I hope the paid version comes with Adobe Creative Cloud Subscription
not sure but it will be integrated to adobe creative ?
may be
no idea
if it goes paied , which im sure it will . that will be the end of it for me
Lol
How much time do i have to wait ?? + What will get me in the mail?
Many of us are hoping it will be included as part of the CC subscription.
evrything is paid atm
right
For me it was One day
oK tHANKS
I have a cc subscription so i hope it will be integrated into it somehow
The time is unknown at this time, they have many requests, all we can do is be patient.
@coral igloo from the #faq Will Adobe Firefly be part of Creative Cloud?
At first, Firefly will be integrated into Adobe Express, Photoshop, Illustrator, and Adobe Experience Manager. The eventual plan is to bring it to Creative Cloud, Document Cloud, and Experience Cloud applications where content is created and modified.
well its like i dont do art for a living , or need to do art for anything. so i'd be paying for it .. for the fun of making art. thats it.
They have said that it will be integrated into Express, photoshop, Illustrator etc,
Thanks
That'll be a great feature.
Perfect
@wooden granite the team is working as fast as possible, the most recent information is in #announcements
you could make out of it as well
mosf of all the art generation is paied for. well anything good
ok bro
You should think of a tik tok profile promoting your art and then sell merchandise with those designs
Who do you think is better? ( Midjourney or Adobe AI ) ?
Thank you for the update.
lmao maybe. currently im not a content creator or anything. i have almost no use for anything i make
tho i could prollly start making websites .. couldnt before cause i was bad at art
try cute cats, you will go viral
lmao
nice
this is not the text tool, this is the image model
i asked it for a country music station logo called "ROOT BEER" and included the text twice

i'd just like to use to to try it out so bad
cute cat --- incoming
right
hiii
next thing you know all the guru will start selling courses on how to write the good prompt
oh im sure
$997
ppl be already making prompt bots
that one is good
But to be honest, Firefly is by far the best and easy tool I have come across so far
Adobe has lived upto its expectations
I hope the add it as part of creative cloud suite, coz I have got one
like the other adobe products
I'm looking forward to just trying out the sketch to color workflow. Could speed up inking or maybe even do flats for me? Need just flats often. Or I can use how it colors my sketch initially and make finish touches..etc.
What kind of delay are people seeing on confirmation emails? Like, over 24hrs?
They have said it will be integrated into express, photoshop and illustrator etc
+48 hours and not confirmation email 
keep an eye on promotion tab if you are using gmail, for me it was there
dang
Same, was in my promotions tab
which is easy to miss if you don't check regularly
not in the updates tab
@final osprey okay that makes me feel better and worse at the same time. Haha
i imagine it should also be there
Check promotions tab in gmail
Hey
probably it has been asked a lot but how long does it take to get access after sending the request?
24 hrs for me
I'm curious what the cost for me will be as I don't use the desktop versions anymore. I work 100% iPad. So I switched from the CC everything subscription $50+ to the mobile app bundle for $15 instead. Will there be an additional subscription fee for Firefly for people like me?
no email haha
Same 48 hours and no email.
I wonder if existing CC subscribers get priority?
They don't.
To me it seems pretty random. I have seen over the last 24 hours some people saying they got access granted within half an hour an I've seen people saying they have been waiting 48+ hours and have not heard. Just confirming, you do get an email when you are granted access, right?
yep, u get a email with access
Just need to wait ๐
what are the prompts plesae , like /imagine or create
I just signed up. I sure many of you will get access before I do. Looking forward to giving it a shot. The tech they're working on is exciting.
try cute cat with blue headphone, studio lighting, ultra hd, photorealistic
you will get something like this
It doesn't work here, first you need to request Beta Access
How did you all got access to it? Already requested, but waiting on it.
Request it on https://firefly.adobe.com/ and wait ๐
it should take less than 24 hours
not sure that I got your question, can you elaborate?
there is no time estimated, approving is happening on rolling basics
they're asking how they can use firefly, but it is not a discord bot
Thanks I will
Gotta wait then.
hi
Hey Adobe crew, its now been around 48 hours since I put in my request for access, but I've still not gotten a confirmation email saying that I put it in. Is that normal, or should I reapply?
no way around
I just saw that it says, please enter a valid Adobe ID as an email address. I wonder if you can only try the beta if you already have the CC suite
It is normal, its due to huge interest. No need for re-requesting
HAHAHAHAHAHA
oh my GOD this was such a pain with midjourney and you literally could not do it in stablediff or dalle
im standing on my desk and clapping
You need an Adobe ID, but that is free to signup for.
look at her hands on 3 pic
I have had the same Adobe ID for 15 years ;p
she got 3rd hand!
buddy do you think i care about the hands
@astral jolt do you have any idea if they are going to switch to a premium subscription? Or will it be free for all those who own an adobe subscription
the hands take care of themselves with more data, the text in the non-text model is 
thank you for the heads up. The email I use for my Adobe ID is an old yahoo address and I don't use it for anything else except spam
@warm wadi We don't know but it's eventually going to be integrated in apps like Adobe Express and Photoshop so I can't imagine them charging extra when you are already paying. But maybe for some features. We just don't know yet.
Makes sense, kinda like adobe stock thing
how do i create in midjourney? Newbie here.
slash imagine then enter
Mid has its own guide for newbies lol
No, you can have Adobe Id without having the adobe creative cloud, as far as I am informed
(not made by FireFly but with SD)
got it
yeah i am very consistently getting legible text on a first prompt with the non-text model that is nutso
At the moment Beta it is free, its cloud based and there are no premium or any other subscription. Once they release it the fact might change
@warm wadi Yes, Firefly is way more than just image generation. I can see text effects, recolor vectors, inpainting, extend image ect to be in the apps included with subscription, but something like text to image maybe you get a certain amount per month, similar to how much cloud storage you get per month.
you cannot generate images writing prompts in the discord, please read the #faq to learn more about firefly
man over a horse
Dude, look a at the message directly above you.
read my post above
is it free?
this is like in 4K Xd
i don't think any of the vector stuff or in/outpainting is there yet
for now yes
Sorry to bug ya Katarina, i thought u were part of the adobe team
I'm very excited for that stuff.
for now. but it won't be free once it's out of beta.
what about commercial use
It'll be available once beta is over.
No commercial use in beta, after yes
And don't try to remove the watermark either.
what would happen then?
what will be the subscription price around?
@vapid arch from the #faq "Can I share Adobe Firefly images outside of Discord?
Yes! You are welcome to share what you create in the Firefly Beta on social. We encourage you to use #AdobeFirefly. Note the output in the beta period is not yet for commercial use."
they will eventually add in the Creative Cloud subscription
Well, this is impressive! Good job Adobe
@austere bough @carmine vortex @vapid arch the beta right now is free. That's the only concrete information right now, we will know more when firefly leaves the beta stage
ik
I think that there is some kind of agreement you are confirming once they approve you for beta, they will ask you not to violate it
yes, but I mean more what if you would ignore that are there any like safeguards in place to prevent it from happening? (just being curious here)
oh crap ive just been screencapping from CTRL+SHIFT+S because its quicker
guess i should stop doing that
pls dont ban me i didn't know 
I do not know the ways ๐
yes, when sharing firefly beta generated work, please always use the watermarked version that you can get via the download
Somebody with information please tell me. Is this firefly based on Stable Diffusion? Because images have the Stable feeling. Can we also train the firefly model with our own pictures?
?bam @shut dew violating ToS
Firefly does not run on SD. We may allow artists to train their own models in the future.
"As long as they are not used for commercial purposes ๐" caity.psd
that was on our own generated imgs anyway
can we see past gallery submissions anywhere?
i wanna download a version of that with the watermark and stuff
Firefly uses images from adobeโs line of stock images and copyright free photos
There's no history just yet, but you may be able to use the back button on your browser (or your browser history) to find it
got it
i'm super-impressed
this was hard as hell to do on midjourney, and impossible in any other TTI
it was a one-shot here
Text has been quite a surprise to me
Not sure what the team did differently here, but it's working!
yeah the base model seems to be able to reliably get 2-3 words on a first roll
i'm experimenting with v-rolls now and the slider is tasty
i don't think it's ever gonna resolve "THE QUICK BROWN FOX JUMPS OVER THE LAZY DOG" but no one ever accused me of setting modest goals
Hello everybody im so curious to try so many ideas
hands to
I tried a lot to use Firefly but failed,
SD isn't trained on language, this one is probably
Were you able to request access?
nor is MJ they're purely trained for their imgs
Yeah I clicked there and filled a form
Protip: the file name contains the prompt used to create the image. It's helpful if you want tips on good prompts.
Perfect! The team is working through the requests and you should receive an email when you're in
Hopefully! Thank you โค๏ธ
Do you have an idea on how many hours are required for invitation? I requested access since the live discussion on Tuesday
Olรก
I did the same and Iโm still waiting. Youโre just gonna get a canned response.
hello all
I have to say that the most exciting point of all for me is the clear commercial use orientation. Do we have any feedback on pricing?
Canned response ๐: The team is working through the requests as quickly as possible and are sending them out in waves.
it can abstract lyric prompts that is great
Nothing on pricing yet
"I bring you precious contraband, and ancient tales from distant lands; Of conquerors, and concubines, and conjurers from darker times; Betrayal and conspiracy, sacrilege and heresy; But I feel alright, I feel alright tonight"
There it is!
horse. horse. horse. horse.
I wish I had something more, but that's what's going on behind the scenes
im giving this four stars so far
Wow
again, this is something i've only ever seen from midjourney, dalle2 and stablediffusion (stock) both just return garbage text when you feed them prose
Is that one horse mentioned four times for emphasis, or the four horsemen of the apocalypse?
lets see if i can find a good shakespeare quote that doesn't run afoul of the filter
Lol
the four AI of the apocalypse: language models, image models, audio models, and the quantum computer living inside a robot to run them all
haha nice
Please teach me. i join firefly to use
Generative AI right now is growing exponentially. Very exciting times.
create a cat dancing
you cannot generate images writing prompts in the discord, please read the #faq to learn more about firefly
hmmm, it'll accept a line from the st crispin's day speech but the output isn't very good (its mostly just hooking on "gentlemen in england")
a generation of people who "know how to do things" will do amazing things with AI. but AI will create a generation of people who don't know how to do things after that.
Are some people not getting in the beta? I registered pretty early on and thought I would of gotten my invite by now but still nothing sadly. Just wondering if everyone will have the chance to get in and beta test
apparently totally random
It must be, Shame. I am a paid subscriber with Midjourney and love it, but love the differences Firefly could potentially bring so look forward to getting my hands on it
No ones knows. Could be less than a day, could be a few days.
its obviously not random guys, just relax
doesn't matter if it is or not, honestly
They cherry pick the youtubers and other people who they want to have access first, its a business
It is random. People who have signed up after me are already in. I'm still waiting. Random.
yeah i mean it makes perfect sense
already has been a couple "i'm a big important influencers" come in here and get laughed off
people signing up after you and getting in does not equate to random. It can still be chosen people clearly.
It is more likl.ey cherry picked based on the answers you give it upon registering, things like What do you use AI for etc
i'm not a youtuber and have like a couple hundred twitter followers, and also don't have any kind of active adobe sub, if they got to me, i'm confident it's either random or sequential but not cherry-picked
this seems most likely
Hey friends! Sorry, been a bit busy over here
dude you make perfect sense to pick early
There is no time estimate, they use logic of rolling basic
everyone hanging in there? ๐
They will use answers to test different aspects
i did answer a lot of those questions but thats because i'm slightly obsessed
Well that's still nonsense. I've been a professional artist for 25 years, and an Adobe user for literally ALL of that time.
Just from being around midjourney the lasy year, you're a presence in the space
like I'm saying from what I've seen of your contributions and activity
getting impatient ๐ซค
Surviving my friend, surviving lol
So if I was someone picking I would have picked you over many other randos
Relax fireflies we will all get a beta invite soon.
Don't mind me, I am not trying to add on, I just hope everyone would have just much understanding for everyone's effort
lol
My question wasn't me being rude, or not understanding, I understand there is a process etc, was just curious that is all.
I think they got a huge surge of people applying, so they had to be more selective than maybe they initially wanted to be, thats my guess
I still think they should prioritize professionals and long-time subscribers if that's the case.
RIP - Stock Photography
Yeah of course, AI is exploding especially when big boys like Adobe get involved it brings up a new breed of enthusiasts!
And if that is the case, just admit it, instead of telling us the same line over and over
yeah I feel the most disappointed about that
We know you're all excited, we're approving applicants as quickly as possible and they may not always be completed in chronological order. Please be patient with us as we work to get you into the beta.
They can't prioritise that, that would be the worst thing to do if you want to ensure your business looks good. and I am also a long time subscriber saying that lol
Check the promotion tab in gmail, that is where the approval mail will land.
Then stop telling people it's first come first served.
yeah paying subscriber doesn't equate to good spokesperson
Exactly!
some of us long-time paying subscribers are the ones who dish out the most heat on adobe lol
lol that's probably true
haha, yeah like Photoshop please fix your reset preference crash when running multiple printers haha
@smoky thistle The process is a bit confusing. For example I signed up for Hubspot AI content assistant tool with my email. I immediately got an email saying my email address wasn't associated with a Hubspot acount with a link to create one as it is required for the beta. I then create a Hubspot account and get another email saying congrats your on the waitlist for Hubspot content assistant tool. With this Adobe beta I have received zero emails, so the waiting is worse cause you begin to doubt if you are even on the list.
hang in there! โค๏ธ
They knew I was going to just try to make it do halftones and then show how it can't yet and then I still have a reason to exist lol
I never mentioned anyone being rude. I asked for compassion ๐คทโโ๏ธ
yeah but also if i'm known for anything on the midjourney discord its being a big pain in the butt so they definitely aren't penalizing that
please visit https://firefly.adobe.com/ to request access, and read the #faq for additional information @knotty narwhal
@terse raptor we are working through systems on our end, so the information we're sharing is shifting. This is our attempt to be transparent > the acceptances is occurring on a rolling basis, but it isn't ironclad ๐
I know I know, I just wanted to make sure no one thought I was coming across rude as sometimes text can be read out of context.
I think you're a consistent power-user and that holds a lot of weight with software
when will it be available to me?
@wooden granite speaks a lot of sense lol
how many days of the week ?
we need an AI text-to-tone-of-voice system to make things sound more gentle and sweet in text conversations lol
The mods are processing as fast as they can, jave a look at #faq in the mean time to get a better idea.
just let a bot send these imgs every 5 or 2 minute or something Xd
Have you've registered yet?
We have all probably registered 5 times at least.
I've only registered once. I was afraid to do it again and mess something up lol
I only did it twice I think, cuz I didn't remember if I finished it completely, since they didnt send any confirmation of the application.
practically there is no need for registering more than once
man I absolutely love Midjourney, but the stuff coming out of Firefly is unreal at such an early stage in development
hello Adobe team! as a paying user of other tools, how long should I expect to wait to get access to the beta? out of all the ai tools, this looks the most fluid. Amazing work. i'm drooling
I have already signed up, I don't know when it will be in my possession, that's the question
if possible, please only register once (I know this advice probably comes too late now). because multiple registrations mean more work for the people approving the applications. that means it takes longer for everyone
I got this when I signed up, so I'm still assuming that means it worked and I'm in the queue.
and you would be correct
I've nearly done it again...a few times...but I've walked myself away from the cliff each time lol
Tim do we know what kinda wait time there is gonna be?
Some of us have bad short-term memory, and things like confirmation emails have contributed further to that lol.
there is no time estimated at this point, they are approving on rolling basis
haha, I know how exciting it is ๐ but please don't drool too much here. someone has to clean all that up ๐
I have already signed up, I don't know when it will be in my possession, that's the question
And many of us are creative professionals which means we tend to be moody ๐
Not proud of it, but it's true haha
So the one problem I see now is that if you're really trying to generate images with that AI-esque vague watermark showing up, firefly won't be able to do that. lol
Hi
There are a lot of applications and we are working around the clock to approve, support and answer your questions. Unfortunately, we don't have a time estimate. However, it can be worth checking the #announcements because if we unlock bigger waves of users, we'll let you know ๐
When this was launched? Today?
it was launched two days ago
Oh I see, fascinating ๐
you coulda just done what U2 did a few years ago, give everyone who subscribes access without them asking... that worked out well
launched as in: the beta was announced
good idea in theory, but with the amount of people applying, the website probably couldn't handle all the prompts. right now Firefly is super fast and that's a good thing. I'm sure you can think of certain other Ai tools which are going down frequently
you know, i love the foresight that Adobe have had on this, this is easily a blockbuster/netflix moment industry wide
that's why we have this discord community ๐ - while you wait you can check out #archived-share-your-work-3-23-to-11-23 or simply have a chat with other fellow creatives ๐
is anyone in the beta able to run a prompt for me and share the result privatley? If this isn't alowed please just let me know
That's good insight. But eventually there will be a ton of people using it. I'm guessing you guys have a plan to ramp up the servers to be able to support the demand?
I don't work on the infrastructure, so my guess is as good as yours there. But I suppose it would be a lot easier to scale up gradually as the demand rises
@pine stirrup I mean firefly is running on Nvidia Picasso cloud right? It can handle it :). Let us in!!!!
alright, everyone! ryan said it would be fine! if it breaks, you know who to blame ๐
lol
lol ๐
all jokes aside, if you have feedback regarding the application/approval process, we're here to listen and relay it to the teams.
I'll receive my invitation to the beta when the product will launch haha
will adobe firefly be free?
Depends on the context you want to use the image in, for example some images are full body poses, and some are cropped, and "better" is subjective, so what I think is better might be different than someone else, just keep exploring until you land on something you love @royal lotus
I doubt that very much
Yes, enjoy now that is free
I am hoping it will become a part of the CC I already pay for
Yeah I think a lot of us are hoping for that!
These things cost so much $$$ to develop and sustain, they couldn't possibly make it free lol
I have to imagine (complete speculation) that the integrated tools that will eventually come to Photoshop will be a native part of the application.
I guess too
This makes sense
But its also possible that Firefly integrated into CC could cause CC to cost more
But lets see
That's what i mean
This is why Substance wasnt put into CC
I sincerely hope this is not the case, CC is already expensive in my opinion
Even more
They left Substance out of CC for a reason
true
are seeds a ... thing in this model? one thing you can do on the other TTIs is specify a seed value when you test prompts so you know for sure whether the thing you're doing is actually having an effect and you're not being fooled by the placebo effect of a luckier roll
if that's a thing here, it would be nice to have that option
Has anyone tried adobe firefly??
Yes, I have
yeah its pretty good
Can i see some results
It is by far the best tool I have come across
Guess i will use Firefly + DALL-E for inpainting a bit
Until Firefly gets Inpainting
It is very easy and intuitive
Could you run a prompt for me and share the results to me privatley?
Sure
Thanks, I will send a private DM with the prompt, just want to compare to current results
i'm suddenly getting a ton of these with my prompts
Looks like @cinder dirge broke the servers ๐
Just got access, my Bard and Firefly invites arrived at the same time ๐
it brings me no joy to say it, because openAI could really use some competition, but bard's a disappointment
Also getting this error
hahaa
I'll let the team know
I try not to get caught up in the competition side of things and just appreciate all these releases ๐
Congrats ๐
Meanwhile i didnt even get a single mail in last few hours
XD
Yep, still waiting...
Was here since yesterday
I'm glad to hear how much everyone is enjoying it though.
Yeah Tuesday afternoon for me. Just hit the 48-hour mark, in fact.
I was at work when it was announced
So tuesday, around 22h was when registered for beta
But man every time, literally every time i look up on this channel the first thing i see is a moderator or admin answering for the 1000th time to people asking when and how they can join
The team is actively looking into the loading issues
๐ค Hello.
Its all about the Game!
I see what you did there
You guys should amend the rules to let people know they can't prompt in here, and add a registration link.
uh-oh, i think the whole thing may have tipped over
snaaaaake eyyyyyyes
I knew the machines would come for us all!!!
wut
having the same issue ๐ญ
same here - call NVIDIA server helpline
SMH @cinder dirge
i has listened!
good to see you too man @slim lion ๐
Hopefully the loading issues will be resolved soon. Team is investigating. ๐ง
they just need to reinstall the latest gameready drivers
and download more RAM
s'pose i better get some work done๐ฎ
Thanks for re-updating us after seeing you wrote it like 5 times before LOL
Much appreciated
can someone with access try asking it to generate a 2d spritesheet for (whatever you want), just curious if it can pull off that kind of operation
maybe something like: 8x8 grid of pixel sprites
its down right now so
second that
there's a pixel art preset though so i'll definitely try that
Hi everyone!
right now i understand some of the more clear-cut use cases, in place of stock photos, and various types of what I would call "pure art", but i am wondering how it does at creating what I would call "animation production art", though i'm sure that's not industry terms. being a non-artist myself
@ornate glen heyoo!
Firefly being down is a sign from the Universe to get more coffee. Brb!
I join the beta and suddenly the images stop loading, sorry everyone!
shouldn't have named it after a tv show that got cancelled after 1 season
he asked it to generate an infinitely recursive mendelbrot set and it broke it
I was replying to the Skynet comment.
eeeey I've been accepted! been checking out how Ff interprets my artistic style with various prompts and then the servers went kaput. I think i need more coffee too @slim lion lmao
I'm ready to play... Let's GO!!!
Firefly is down = touch grass
Saaaaaaame! lol
Or apparently drink coffee lol
Besides it's raining here haha
Why not both?
the answer is always coffee โ
if everyone stops using it for exactly 3 minutes, everything will work again
doesnt it tho? lol
I'm just glad that Adobe is in the game.
hi . i cant find how is it work. anybody can help me?
Also, @quaint lily i would like to recommend adding a 'status' channel in discord for admins to post updates on performance and version changes.
Firefly down, Quick call Malcolm Reynolds, Jane, Hoban Washburne and Kaylee
See #faq you would have to register for beta, there is a website link there
Go here: firefly.adobe.com
You need an Adobe ID, and then you need to fill out a form and request access.
how long after you apply for the beta do you have to wait to get access?
this is what i get for "sprite sheet for black mage" with the pixel art preset, so yeah, that's not quite ready for prime time 
ewwww
Hi @late briar : Can you specify what you are having problems with?
heh, i think it probably needs more context to understand what is meant by a sprite sheet
i was do that operation in website and got invitation to discord
now i dont know how asking from this ai for a simple picture !
When I click request access nothing happens can anyone help me?
Thanks for the suggestion! Point taken and considered ๐
Beta applications are being approved on a rolling basis and you'll receive a confirmation email once you've been granted access. At this time we do not have an accurate estimate for when access will be granted. Hang in there!
i'm clicking generate every once in a while to check if its back up and every so often one grid is going through
you will receive an email wit further details
sorry for my weak language . english isnt my native
Until we're all accepted: Just keep putting out milk and cookies for Adobe Claus
Which browser are you on?
Chrome
yw and of course. ๐
and later, I guess it will be added here?: https://status.adobe.com/
Adobe Status provides service availability and performance impact information for all Adobe products and services. It includes incidents and maintenances that are ongoing or in the past that affect Adobe products.
GPT-4 says this is a better prompt to try. might be too big, i dunno.
"Create a 2D, tile-based spritesheet with a pixel art style, suitable for a top-down RPG game similar to those created in RPG Maker. The spritesheet should consist of 16x16 pixel tiles, featuring various terrain, objects, and characters. Please include the following elements:
Grass, dirt, and water terrain tiles with smooth transitions between them.
Forest elements like trees, bushes, and rocks.
Pathways, roads, and bridges to connect different areas.
Structures such as houses, shops, and castles, with exterior and interior tilesets.
Props like signs, fences, and treasure chests.
A variety of characters, including a protagonist, villagers, merchants, and enemies, with walking and idle animations.
Make sure the tiles seamlessly connect with each other, allowing for the creation of diverse and engaging game environments. Use a cohesive color palette and consistent style throughout the entire spritesheet."
We have channels in some other languages as well
GPT-4 does not know how to prompt and it is a mystery to me why everyone thinks it does
isnt the entire point of a sprite sheet getting the same character in different frames
just a one-shot try. clearly it doesn't "understand" the idea and needs some structure, even if what it produces inside that structure isn't precisely right
the problem is that TTIs don't do "instruct" language in the way LLMs do
ah
yes i see the other channels , but im persian and you havent farsi channel ๐ . english is better for me . i learn more...
so imperative commants like "create this" and "it should do this" are just noise
well i defer to your superior knowledge on that.
theres a few things i can grab from there
but i also don't know what the prompt limits are
like is this clip? do we get 77 tokens?
No worries, we are here to help. Happy to have you here.
That's strange. It should work on Chrome. Can you try a different browser, Safari or edge?
I prompted 'a cabin in the woods' and it gave me a beautiful photo of a cabin. Then I added 'with a monkey in the yard'....and it gave me a monkey that looked like it went through the transporter pods in 'The Fly'. I wonder why it had trouble.
maybe its a GAN 
Try a photo
Does "Can't load This image produced an error. Please try again." mean that I've done something wrong?
i do believe i'm nudging it a bit closer to the core concept
but the results are still a bit too wild for actual use
2D spritesheet of pixelated character with red hair wearing red armor in assorted various fighting poses, sprite sheet, pixel art, NES, 8-bit (art, pixel art presets)
so uhh not so much the armor
it looks like the way your prompt is constructed is almost like adding tags to your core prompt
(i'm prompting both spritesheet and sprite sheet because idk how the text stuff works)
yeah, that's the way to do it in most of these TTIs
I had a similar but different take on the cabin theme ๐
i bet i could stack on a few more synonyms
Seemingly can't put any type of weapon in the prompt. Tried to have a rifle and a sword, and got the words violating the terms prompt. So instead I settled with a sausage on their shoulder.
Now I hear you. ๐
Village
A fearsome warrior.
i really hope that gets loosened its gonna really straitjacket a lot of creative work if this can only be used for stuff that passes muster at disneyland
when sharing share with watermark please
I need sleep, forgot how typing works
probably doesn't understand the word spritesheet. If you describe it as a grid, the results will probably be better
*detuned victory theme*
@silk tartan We're aware of the issue #community-chat message
so yeah, the answer to can it do pixel art is "sort of but not really" (this is also the answer for every other current TTI so no shame in that)
Its solved though
its working again
Come for help, be inspired. The Adobe Support Community is the place to ask questions, find answers, learn from experts and share your knowledge. We have everything to help you get started, learn how to's, tips and tricks, and unlock your creativity.
@split pendant now this is getting spicy
that's a really nice...sword...
close enough for the canned reply
hi @charred plover At first, Firefly will be integrated into Adobe Express, Photoshop, Illustrator, and Adobe Experience Manager. The eventual plan is to bring it to Creative Cloud, Document Cloud, and Experience Cloud applications where content is created and modified.
excellent. it's getting there
check out this refreshing beverage
looks like it might be able to be coaxed into doing it with the right amount of prompt design
Sorry Jean should of been more specific it's unrelated to firefly and related to compatibility on a Microsoft Server Citrix farm.
alright its time for me to raid my midjourney archive for text memes and see how many words i can coax out of this
Add Adobe XD to that list. Train it on the best UI optimized designs and be able to prompt it for website or app page designs.
Would be cool to verify signatures on Adobe using AI
Welcome @shut pawn - you need access to the beta before you can use Firefly. Check out #faq for how to apply
Have like a signature database assign it to a token and an ai references a signature check
Block chain ai signature recognition
My prediction. Generated art that will end up in layers in Photoshop.....Bring it Adobe. That will hurt Midjourney.
Would be great to collab with MJ, not hurt them ๐
Dora the explorer
Midjourney would need to change the dataset they train their AI on before they do that though
So scrap everything and start again ๐
nightr dress concept art
okay, i'm definitely getting better results for text when i repeat it in the prompt
which makes sense, midjourney was like that too
sitting here waiting to be accepted
Johnny Depp portrait
Pretty much. Midjourneyโs dataset is filled with copyrighted material used without permission, the very opposite of what adobe has done
Hi @bold loom if you want to learn more about using firefly please read the #faq and visit https://firefly.adobe.com/ to request access, it is not a discord bot
You should get a form to fill out
This chat sometimes
oops I was scrolled up further than I thought..
it doesn't know who uncle sam is, but props on the text again
jack watson told i can share my works here
jack.watson โ ma 21:41-kor
this specific channel is @elder tusk , please post work created with firefly here, I already said you can post your other work in the chat-english channel if you'd like
An Illustration of a kid dreaming milk
Any moderators/developers of firefly here available? I have a question to ask regarding how firefly was trained on a specific element
We'd prefer you didn't, but if you're going to insist on posting..... and please do not spam the chat with multiple images, @elder tusk
sure can you head to #1087561080638808095
This Discord server is dedicated to the cloud-based app called "Firefly". It means that all discussions and content shared in this server should ideally come from Firefly. If you want to get access to the app, you can read the #faq section for more information on how to apply for beta access. Once you're granted access, you can share your work from Firefly in the #archived-share-your-work-3-23-to-11-23 channel.
ok i leave , tons of servers everywhere..
bro this reflection wtf
i don't even want to use the dedicated text model this is awesome
If anyone is interested we are having a live discussion with the team on #AdobeFirefly on Twitter Spaces. You can join in here: https://twitter.com/Adobe/status/1639008892696010762
yeah i cant wait to get access
im coming
Im still waiting beta ๐ญ
same
twitter spaces on right now from Adobe page
a nice landscape
haha just seen it already posted
i'm stoked on the beta!
Me, tooโฆ Till that will be, Iโm organizing ideas what to make in the future))
bruh, i didnt even know Twitter has this function
mild UI complaint: these instructions for the two parts of the text generator bar are damn near invisible
I just sent in my firefly beta request, seeing all of these cloud based machine learning approaches being deployed is sparking a lot of ideas for where we are and where we are going