#💬│general
993158 messages · Page 14 of 994
You can kind of do that by deleting them for now (while they're still in progress)
but that's a good point
Thanks! Glad you do!
even better when i could generate a bloody fightscene without restrictions
Apologies if we are all over the place
Currently it's just me and Yoss on the community side 😅
No need to
This is actually really professional
I'm proud as hell
Thank you! We try our best to be!
Hopefully in the future we'll be able to scale up on the community management side as well
Also you could ask people to help with your generations
like to donate their gpus or sum
so it'll be easier for everybody
also does Ideate use up anything?
Please say no.
ideate?
Ah, that
yeppers
I believe there's 0 restrictions on that 🙂
at least for now
But I find it better to edit the prompts anyway 😅
tru tru
and also yall should add like
a hover thing for the models
like an explanation for em
Oh we do have that in the Finetuned models section in a way
instead of having to tab out to the model page
As for the drop down menu, we're revamping that to an modal type design
so cool

🫂
This is actually
so fricking well done
How long did it take?
Can't answer that unfortunately as I was brought on just right before the early access launch 😅
o
at least 10 minutes 👀
Lmfao
probably more, but can't go wrong with at least 10 minutes
I'm using this thing more than playground ai

also there are pther models too
like midjourney diffusion
yiff model
redshift diffusion
damn the upscaler for the image takin long
the influx of people is massive at moment 😁 I think we're going to see some slowdowns today
still aint workin though
Imma just use photo ai and upscale it on my computer tbh
brb
/ima
.
wow
Images are great. Quick question - Is there a way to get a consistent character or look and feel from generation to generation? Takes alot of patience in Midjourney. If I get a person in an image and I want that 'face' to be in another image, what is the best way to do that?
you can use that image to influence the others but i don't know how much it can help
at that resolution, jpg is fine
is there an option to rate the results?
there's no loss of quality if the image is generated as jpg I believe
using 1024x1024
no it's png
jpg is compressed
Hey! Leonardo currently isn't available as a discord bot so prompts will not work in the discord. Please kindly either wait for the invitation email with the web app link.
it's only compressed if there's something to compress though 👀 if the image is generated as jpg with 100 quality there is no information to lose
Huh that's true
I'm just scared of losing quality
since uh
upscaler not workin rn
Hi
If you have yet to get access to the web application, you can quick intro in #👑│introductions and then taking a look at #✨│priority-early-access for details on getting an invitation to the web app 🙂
quality is solid 💪 I use it for works I sell online and there's never been an issue to due artifacts and such, in fact, you can upscale the result further since it generates further details rather than just upscaling.
I'm using the topaz ai app to upscale rn
Topaz works great
oh yeah, you can combine the two to get 10kx10k~ images fairly easily
my laptop is too weak for topaz unfortunately 👀
Hi all: when I upload an image: how am I able to give some prompts while using that image?
the prompt box at the top of the page
Is it normal for there to be a big white bar on the right hand side of browser window? It takes up approximately 25% of the browser window.
Ohhh : so as long as the photo is in my que at the bottom left, the prompts will alter the image?
seems weird, maybe try refreshing? shift f5 to be sure. The web app doesn't work to well on mobiles btw, since they're developing a standalone app
yes, without prompt magic the AI will draw on top of it, with prompt magic it will take inspiration from your picture, but start drawing a completely new one.
You can determine how much of the original pic to preserve with the init strength parameter below your picture (if you have prompt magic off)
can generate ai from discord like midjourney?
No you can't generate from discord yet.
no, only from the web app
On a Windows 10 PC using Chrome. It took a couple of refreshes but that fixed it. Thanks for the suggestion.
GM! Don't forget, we do have a fun event going on in #1063537670715875539 called the "Sim City" Challenge.
Would love to see some of you enter in and give it a shot! 
Not currently , reinforcement learning is rather new tech in machine learning so perhaps we might implement it in the future when it's more viable
What if you put your negative prompt in the Ideate section
what would happen
would it work or
Absolutely not normal
Do you have any addons that may have an sidebar popup?
Shift+f5 as well and see if it works
That I have no clue. But likely not possible ;0
man.
I wonder what happens, give it a shot!
too many users
We're going to have to scale up things quickly
uh oh
Stop being so useful!
WAAAAAAAAAA
Not sure why there is a Dreamstime watermark on my panda images. Is it because the model used it? That's pretty bad. Every one of them. Or is this a Dreamstime AI?
Sweet, so just gotta wait 24 hours for access, excited to try this out!
can happen, I suggest putting dreamstime in the negative prompts alongside watermarks. It's an issue residue from the stable diffusion 2.1 model.
Just got access! That was fast. Its go time now! Thanks team
Woohoo, welcome!
/image a tall leprechaun pouring wine into a wine glass
You're almost there! Excited for you to try it out
Yes. Im all in. Thank you
That command doesn't work here friend. You'll have to wait to get access first, then let your creativity flow.
Hey there! Please be informed that Leonardo.Ai is currently only available as a web application. /imagine and prompting for image generations do not work in this community server. Thank you for your understanding!
OMG! I just tested my very first prompts in app.Leonardo and I couldn't believe my eyes. I tried several ai image generators including midjourney but this beats them all. So happy with this new release. You got an admirer 🥰
Please do share! #🖼│image-share
Thankyou so much for lighting fast early access
Thank you so much for the compliments on behalf of the team! 
hey
👋
what's upp
Every image is a panda now, from the original prompt I used, and none of the new prompts or models are being used, I just keep generating the same panda over and over and it always has a watermark for Dreamstime.
did you set it as img2img perhaps?
check in the bottom left
you probably have your panda pic right there
I am going back, then using a new prompt. How do I get out of this?
press on the bin on the image in the bottom left to remove it from img2img
Hmmm... OK, I don't understand how it got there and stayed there, but I will watch out for it
Perhaps in the future but I can't give an certain answer at the moment 🙂
yay

hyper realistic, highly detailed digital art masterpiece, smooth cam, de leon eric zener, dramatic pearlescent, soft light, hd 8k, sharp focus
🥹 BRO COME ON
upscaling with leonardo already does miracles for faces though 👀
Fire as hell
@gusty bay the reflections on the car are awesome
it's actually img2img + upscaling technically, but yes, all in one step 👀
⚠️ Hey there! Please be informed that Leonardo.Ai is currently only available as a web application. /imagine and prompting for image generations do not work in this community server. Thank you for your understanding!
If you have yet to get access to the web application, kindly consider making an quick self-introduction in 👑│introductions and then check the ✨│priority-early-access channel for further details. Thank you!
Hi, still not sure what to do, filled out info form, am I suppose to wait for another invite???
Yes! It should be within 24 hours, but with the demand there may be a delay.
Dreamstime logo keeps showing up, very disappointing
are you using "Prompt Magic"?
disable that toggle if so
Not using Prompt Magic here
what;s the prompt you're using?
I'm sorry it's not working out for you, but personally I've had really good experiences on this platform
cute cartoon fawn deer sticker, vector illustration, white border, white background (negative: watermark)
Hello there,,, is there something else to do? I don't know how to access Leonardo.
Thank you, can't wait!
I put it because it kept showing up. The second one has the watermark
Just wait for now, currently we have a large influx of users so processing will take up to 24 hours. Apologies for the delay! 🙇♀️
@summer adder
try setting the guidance scale a bit up
i have it in 10
and use the Leonardo Style
This one. It originally showed up in all but one Panda after multiple generations
OK
the Panda one seems to have used img2img, so of course it would retain the watermark if the original image had it
I dindt get my mail welcome to join the site, he take much time I don't know why
Well, all 4 of the originals had it
@summer adder based in your feed, you used the img2img on the one that had a watermark
happens when you use prompt magic, the 4 images come out basically as variations of each other so if one has it all of them will most likely
the first 3 don't have one
Hi All, Is there a paid plan available yet ?
not yet
This may sound like a stupid question, but can you tell it what size of graphic you want it to create, so you don't have to crop and have it distort the image?
f you have yet to get access to the web application, kindly consider making an quick self-introduction in #👑│introductions and then check the #✨│priority-early-access channel for further details. Thank you! 🙂
img2img? or text to image?
img2img
I think when you upload the image it automatically selects the same width and height?
works for me at least
How long does it take to get the email to join ?? Anyone ??
24h? Faster if you go introduce yourself in #👑│introductions and follow the instructions from there
lots of people joining right now however 😋
Are we allowed to post images here?
#🖼│image-share (is the best place)
Yeah, I don't know how to get rid of this watermark...
As you can see, it is in 3 of 4, the fourth was not up to par.
You have already used the negative prompt 🙂
hi where di we go to start making images
I did, I took it off, it doesn't matter. I can even use a different model. I think it might be the Cute Animals style?
Gotta go eat, later.
You have to wait for your invitation in your email because of the demand your application may be delayed, but rest assured that you will get it.
Try using the AI canvas to clean up the watermark instead if you really have to
Actuallly, no, I was using Leonardo style there
thank you for response
community models may work better in that regard. Some of the Leonardo models pull too much data from SD sometimes
Is anyone having problems with upscaling at the moment?
Nope, what problem are you running into?
gm, how do i get early access sirs
As we are currently experiencing an large influx of new users, please be informed that invitations will currently take up to 24 hours to be sent out. Our apologies for the wait and thank you in advance for your patience!
ah as in in the emails ?
Yes 🙂
It says it's working but then it never seems to upscale it. I've tried refreshing a few times too
wait u can see nsfw here?
Currently likely to be an issue due to the influx of users
We're looking to scale up on things asap
is this the email frappy @midnight cloud
Not exactly. They do not show up by default on the community feed, but you can generate nsfw images yes
u indo ?
That's just the confirmation email to let you know that you've signed up for our mailing list 🙂
much thanks ayu
And no, though I've been asked that sometimes!
I am having a trouble with upscaling
Would it be an delay / non response for upscaling?
will this also run in discord or just the app?
OMG... you can train your own model WOOOOOW... give me the paid subsription... I SIGN 🙂

Just the web app.
i saw some youtubers run it in their website
For now it's just a web application, we have Discord contemplated, but that's more in the future.
Hey can i ask a question, can i get an ai version of my image
i mean image to image is this included
Through the app, definitely.
do we create from discord like midjourney/blue willow
no, from the web app! #🖼│image-share message
wait wrong link
this one^
thank you so we have to wait for the email or can we visit the app
hi again, how can i buy a liscence for my images created
Yeah the invitation email will contain an link to the app as well
You don't need to, all rights belong to the user who generated them 🙂
ok thanks again
You have to get a license
please dont let at be a free use forever...
oh really, i appreciate that
Read section 3 of our Terms of Service in regards to ownership of images: https://leonardo.ai/terms-of-service/
👍
You're welcome! Do read the TOS regardless, would be informative 🙂
excited to see what Leonardo AI is capable of
thanks
Model Promotion. If you train a public model that is popular on the Website, we reserve the right in our sole discretion to promote the model to permanent status on the Website, which means the promoted model will no longer count towards your Model Allowance and cannot be deleted by you. Users have the ability to keep models private if they choose if they prefer not to be subject to this Model Promotion clause.
keeping a model private is only for paid users later?
yeah i will
Models are actually private by default
ok cool
And will continue to remain private. It's just a clause if an model is made public and used by an large number of people
ok great... is your API free to use?
Welcome to our community! Do consider making an quick intro in #👑│introductions and then taking a look at the automated bot response for more details 🙂
How can i start created images in this App 😂
OMG i cant believe this!!! i cant wait to use it
looks promising
Currently for testing, yes
But will be part of paid plans in the future
Send me prompt for image creating🤗
Welcome!Leonardo.Ai is currently an web application that's not part of discord, you'd require an invitation to use it. Do consider making an quick introduction in #👑│introductions and then following the automated response from the bot for the rest of the steps 🙂
To me or?
Thnks
A pleasure 🙂
Very exciting
Look what I've just generated using Leonardo 😍
Really nice!
Thank you 😊☺️
Its amazing really
Thank you 😄
What channel is best to ask questions? I'm having issues with the img2img option. It doesn't take my prompt into consideration
you should probably lower the init strength, you can find this option in the bottom left of the page, below your image
new to the discord, but that's nice that you can adjust strengths of images vs prompting. looking forward to testing it out.
Cool cat
Welcome to our community! As Leonardo.Ai is it's own web application separate from discord , image generation is not available though here. Do consider making an quick intro in #👑│introductions and following the automated response from the bot for more information on getting access 🙂
Hello just one question how do you generate stuff
#1058409094421155870 message check this out
Can I make modifications on a specific seed number?
Welcome new members!
yes, under advanced options in the bottom left. Set the seed and turn on the fixed seed!
Hello there! 👋
Hey there! Welcome!
Thank you!
guys how much does it cost to generate art?
Hi there. I'm newbie and I would like to use this amazing Ai
how do i use the ai
Not gonna lie, this server gives off a friendly vibes 
Im pretty sure you need to get whitelisted first and then login to the website
correct!
so i need to pay?
no, it's free! #1058409094421155870 message
I heard its free
My prices range from 15-100$ depending on what you want :3
they will introduce paid tiers later on, but maintain a free tier! Currently there's no paid options, it's only free
How do I use this tool?
#1058409094421155870 message read here
You should introduce yourself in #👑│introductions and then you can get priority whitelist in #✨│priority-early-access
Can't wait looks so good 🙂
True, it looks stunning
It's 1:49AM and i can't sleep because im too excited 
This tool is a game changer in the way you can create and use your custom models! It would be awesome to be able to mix models with your custom model when prompting.
how long will it take for them to get my invite
Because of the large influx of people registering it would take up to 24 hours.
why do they whitelist and not open it completely?
Because this is an Early Access?
oh so it will be paid in the future?
Hello, web dev and DnD enthusiast checking in :D
and its an early testing
when the app officially launches it will be completely open.
Vecna said there'll be Paid Plan and Free Plan
is vecna a mod?
Vecna is an Early Contributor
not a mod, but I'm pretty close to the team and I talk to them often! 😁
A monster...
Possibility.
Such a good movie.
@spark cosmos is secretly the mastermind of the whole project.
I wish there'll be more emojis added soon xD
waiting for them to add these 👀
need to work on the emotions though

I don't know why but whenever i type i need to add emoji to not make it feel weird 😂
It's just that it's prettier hahaha
looks nice 😮
Your name stands out the most for me because it means Beautiful in Javanese 🤣
Thank you ! I've actually been told that by another user from Indonesia, it is however part of my first name which is Japanese 🙂
mfw I read the word "/imagine" for 256th time today
👍
Maybe add a autorespond for /imagine
trying to understand if it's a real pic or generated 👀
there is actually, usually people ignore it lol
we did 🙂
totally
very well represented
/imagine not having to use /imagine
@neat axle nice
/imagine a pc that is better than my current one spawned in front of me
best processor ever 😢
hey, that's why we use web apps and not our pcs 😁
That's why im looking for alternatives so my pc wont get roasted 🤣
I usually use Google Colab to run Stable Diffusion tho
same same
honestly I use it all atm, MJ, leonardo, SD, GIMP, paint
more is better
How do I know if I made it to get the early access?
You will get an email, im pretty sure.
Google Colab moment.
I have a pro account since I needed it for my own ML research 🤓
so I didn't need to do that haha
The misery of a broke student 👍
How do I generate an image?
You can generate an Image from the website.
But to do that you need to get whitelisted first.
Is it like midjourney, where I put /imagine?
Not yet.
Oh okay. You don't make the images on discord?
Yep
You make it on their website?
Yes
Thank you for the info. Can you link me the site?
Hello everyone
Totally.
Thank you
Hello there 👋
How it's work?
Make sure to get registered first, and it will take up to 24 hours to get whitelisted because of the large influx of newcomers.
Well this AI can generate an image according to your prompts and if you don't know how to write the prompts you can see the gallery in the website and try other's prompt to generate the image you like.
You can access the AI by registering from their website.
You can wait up to 24 hours to get whitelisted or you can get Priority Whitelist by introducing yourself in #👑│introductions and go to #✨│priority-early-access and fill the form there to get Priority Whitelist.
Okay.
Thanks 💜
You are welcome 
pro explanations
Do you think so?
absolutely, top notch
happens to the best of us
you're in the perfect environment to thrive though
we're all prompters here
Honestly i myself dont know how to make good prompts but i usually just take some prompts from others and edit it to my liking 
Can't wait to try the AI tho cause it looks so stunning xD
good artists copy, great artists steal
Both are me.
Btw what is this Courtier role?
the role is given automatically after you write an introduction in #👑│introductions
I see not everyone gets it but i forgot that it was because there are too many people typing all at once 🤣
Welcome new members. 

😄
hi, im a comic book artist. love ai, and would love to try this one. cheers
You can introduce yourself in #👑│introductions to gain Courtier role and you can access #✨│priority-early-access after that, by filling the form in that channel you can gain Priority Whitelist for LeonardoAI.
Hi there! 
What do i do now?
If you already registered yourself in the Website you can introduce yourself in #👑│introductions and get access to #✨│priority-early-access then fill out the form there to get Priority Whitelist.
Hello everyone 👋
👋 Hello Hello
Welcome new members!
so is the email just to verify your email ?
helloooo
Hey, i just noticed that in #👑│introductions the bot doesnt work properly because of the message submitted by the newcomers are too much, maybe it'll get solved by making another introductions channel?
not sure tho
I think your email is used to inform you when you gain your access and will be used to login to the website.
It's because we're getting an influx of members. It's just doubling the message because there too many people. Should correct itself soon lol I hope. Thanks for the heads up friend.

great to see so many people signing up!
I watched a YT vid of someone using this tool - looks promising!
Glad you are here too!
so when you make a image is it private or public
Im not sure about that but i think it's public, but again im not sure because i hasn't gotten the chance to try it out.
public, you can ask for your output to be made private if you have professional needs
Just want to say... Leonardo is absolutely amazing..I've been messing around with it for hours.
Good to see that you are enjoying it, can't wait for me to try it too xD
got any prompts you want me to give a shot to?
got a little time while food cooks
Don't think so
My brain couldn't
haha alright
Someone really tried to generate Images in #1053969672699129856 🤣
Thanks for sticking around everyone!
We will be getting the invitations sorted eventually!
Just waiting on somethings at the moment on the backend 🙂
Much apologies for keeping you all waiting 🙇♀️
Im pretty sure that somethings are snacks

I was holding my laugh cause i don't want to look like an Idiot laughing in 3AM 🤣
ViralHog indeed.
will there be a cost after early access? If so how much?
There'll be Free Plan and Paid Plan
Im not completely sure about the cost though.
Looking forward to trying the AI soon as I get access! Already do a lot of AI in different engines. Did the Intro and Priority thing so just waiting. Hi! 👋
👍
Hello! I hope you have the patience to wait because there are influx of newcomers here.
Yes, although we have plans to retain a free tier. As for costs we are unable to provide an estimate or exact amount at the present moment 🙂
model training is free?
Im pretty sure its free
Yes!
No problem! I'll wait - I'm eager to try this engine but I'll be patient!
Welcome and apologies for the wait! We're just sorting a lot of things in the background, but rest assured we will be sending out the invitations when we are able to!
Free with a limit of 10 per month currently
Me with no experience using Model Training be like : That's cool
use like 10 photos of yourself
I roughly know how to train the model but I don't know why I am not interested in doing that at the moment.
i will make my dog maybe
Only if you're really wanting an certain style or subject matter
We got plenty of models, both official and ported and also community ones 🙂
Mhm...
Should be plenty to play around with
That's true
Recommedations to start are RPG 4.0, deliberate 1.1
Dreamshaper does pretty well too at times
Ayu you trained this?
leonardo select is the best!
each model says whether it was trained on SD 1.5 or 2.1, but they have custom solutions built on top which are very impressive
Can we download the models that we trained?
I am unable to upscale any of my images, what could be the reason?
unfortunately no
because there's too many people coming in at the moment, servers are struggling hard! Best to focus on normal generations for now I'm afraid
Website is very busy at the moment. Flood of new people
So I submitted the form then exited the app will I still be whitelisted
Ohh understood, i notice that generation takes lots of time during these hours, but this is something new
When you are in the Whitelisted List then you can use your email to re-login
hello hello
Leonardo looks really interesting. The UI is clean from what I've seen.
This is running SD in the background right?
Yes, i think all models are trained on SD of various versions
If you have filled out the form, you should expect your invitation in your email address
usually creation takes 10-20 seconds, and upscaling takes 30. Now that the userbase is exploding we'll see 👀 I heard they were upscaling their backend
Glad to see they are responsive to the growing userbase.
We are all waiting for the cookies
Can't wait to see some crazy new features with these upgrades
lots to look forward to, from what I heard!
I'm looking to create scenes and graphics to enhance my photography.
#1058409094421155870 message read this to get quicker access
Just ran through the process for faster early access. Wondering if it's the kind of thing that could take days or weeks from this point, or if its still fairly quick?
It was pretty quick, i didn't do anything extra other than applying on the website
okay good to know. Thanks!
But because there's large influx of newcomers it might take up to 24 hours, hope you can wait 

If its money, then i'll gladly take your offer.

I'm sure the team will hire you as a community manager 👀
personally I just want to generate images foreeeeeeeeeeeeever
@civic violet for president
Well then i'll bother you with some random things so you can't do it 🤣
uploading "dataset images" to a custom dataset is giving the AI info to make similar things? or what does it do
hi
who owns this? is Bil Gatjs?
Who is Bil Gatjs ;-;
Is there a limit to the pictures?
microsoft guy
I think so, it can make a keyword that can be put into your prompt to it'll make image based on your dataset
I think this is owned by Elom Tusk
are you familiar with dreambooth? I think it's similar. It trains the AI with the images you show it
yes, introduced today
ok so its secret?
It's totally a secret
Are there privileges for earl users? Is it limited for everyone?
trolling?
not yet but Ayu is fighting for our rights 👀
i just joined the discrod ... whats next ? how do i get access to the ui ?
Bit of a bummer on the limit, getting an image just right can really burn through those credits
@analog nebula 🫡
https://www.linkedin.com/company/leonardo-ai/?originalSubdomain=au maybe you'll find your answer here 👀
That's true
Nah it's owned by Elom Tusk 🤣
btw the limit is 1k/day right?
ty Sir
the limit is daily but I'm not sure how much, I seem to only have 200 👀
👀
Frankly I agree - IF this server can be comparable and/or better than Midjourney AND be less strict on words and content... I don't mean allow XXX stuff but you can't even use the term "Blood" there. If we want to make scary Beksinski styled art or try for attractive if not X stuff (say a lady in a bikini, more meant to be a Vaporwave album) the term "Art" needs to apply.
anyone know why upscaling does not work?
Too much users so the server couldn't handle it atm
No these are the platform models, the defaults 🙂
honestly, their upscaling is some real magic
I sometimes take midjourney images, put them through the upscaler on leonardo and boom, we have a masterpiece
Unfortunately we needed to find a balance between amount of users we can let in and our computational power limits
ok I will test it
I absolutely hear that of course.
Their first upscale is great - but then any further ones make the face go melty or missing. I use Gigapixel after the first one, doing some edits with Affinity Photo 2 for hands/odd faces, etc.
you're talking about which upscaler? I'm talking about leonardo's
it's the finishing touch I use for most of my pictures
My pc couldn't handle Gigapixel 🤣
i regret buying gigapixel
Do consider making an quick intro in #👑│introductions , then read the automated message from the bot for details
don't regret it! Apply gigapixel after leonardo's upscale and enjoy 10k images with no artifacts 👀
what is used as upscaler here?
some sort of diffusion model, it actually generates new details!
real esrgan servs me better
nothing prohibits you from applying real esrgan after leonardo's upscale, it's two different approaches that work very well together
So many people talk it up... Whats the regret?
I use Channer for mine
terrible results compared to free models
ATTENTION COMMUNITY! 
Don't forget about:
The "Sim City" challenge is still on, join us at #1063537670715875539
(All the instructions can be found there).
lol this scared me
I thought frap was going to arrest me for generating too many images 😔
Personaly I LOVE Gigapixel. I did wait for sales and got their other stuff (save video) but I was able to get classic low resolution pictures and smooth them and upscale them and they rivalled modern high gloss and megapixel era photos. Definitely I'd be in trouble for posting examples here, say "Men's Interest" mags from the 90s and/or uploads to the web when it was looser but we operated in upper kilobytes/early megabytes so a 40K pic was big. Now I'm talking about buying them <6 months ago. Earlier the upscalers didn't have modern AI and so the software just made the pic larger and blurred the dots diagonally with no real adaptive filter. So if you tried them 1 year ish ago you might try again.
Hey, I'm there lol
lolol
I am mildly curious of what constitutes an "early contributor"
I mean is a date joined thing? Or???
no! Just be active, helpful, post images, models, help the community grow here and on socials and you will be an early contributor too!
any of these actions can qualify you, personally I just help around the server a bunch
Totally won't be me 🤣
Ahhhhhhhhhhhhhh. Ok. Fair enough. I've added some images, but models just are not something I feel like I can do especially well.
I could never arrest you, you're too helpful lol
🫂
Never throw anything that can still be used even for a tiny bit. 🤣
Hello
Hello there!
Please go to #👑│introductions to make an introduction and you can read the automated response by J.A.R.V.I.S for further details, if you haven't done that! 
hello!
hi!
Hello there, Sophia!
bruh it went to 14k to 18k in just like 2 hours
Server models are good on most upscalers - but they re-re-re render them so they change the image a bit. Sometimes its good as it can fix the bad anatomy and other artefacts but sometimes it adds them. Or it adds the "Overworked" hazard that RL artists get if they try to make a painting look good and take weeks, months, years tweaking it. So I usually go Thumbnail >> first enhance >> Gigapixel upscale for things I'll use and publish. At the mid range to high rez/printing Gigapixel really gets the enhancement done. I was originally just satisfied with the first Midjourney enhancement but some associates in the TTRPG world wanted pics I made for them to publish in RL print enhanced. So I went around to trying the new upscalers using modern AI and LOVED it. I'd tried those 1.5 years ago and they weren't worth the time to play with but again anyone should try unless they have an ancient and/or budget computer. The latter subscribing to an uploader IF they need print quality might be worth it. Web and retrogaming, fine as is.
yeah, the image definitely changes with the current upscaler, but I love the result personally
got access!
That's good!
Is upscaling not working for anyone else
We are GROWING
Facts
wow nsfw is unlocked?
Its fun to see how the upscaler decide what details to create
yes, too many new people currently! Take a deep breath while the team works on upscaling the backend
I think that's because the heavy burden caused by large influx of new users so the server couldn't handle it or smth like that
Oh
I got access just 5 mins ago.
Pretty fast ngl
Got access in less than 3 hours
Upscaling is working but there's a queue !
Also we'll be pushing some updates to the upscaler early next week to improve its performance with some models.
Its waiting time 🤣
well I'm sad no more unlimited renders but also very understandable, had a lot of fun these past few days tho
what upscaler you use
that's an old message, sorry, check back in 5 mins 🙂
lolol thanks
our own pipeline
Looks like, I got a lot of NSFW images on the feed today

Please check #📢│announcements if you haven't already. Will be some downtime on our site 🙂
Apparently the time of unlimited rendering is also over. I'm being assigned tokens now.
we're improving filtering - NSFW shouldn't show on main feed
is this something you fixed? earlier it was just aborting for me
tokens are a good thing probably, render time was getting sooo high lots of people probably using
Hm, it shouldn't have been aborting
looks amazing
any team members here?
happened around 8 times in a row, I'm pretty confident about it
both
the results? hahahah
Yeah, I've noticed that, but only within the last few hours. Rendering was usually super-fast. But tonight it was slow, and now there are tokens.
Our inference queue was getting hit hard, so we had to bring tokens in sooner than anticipated. And this was after we added another few clusters of GPUs 😂
🤣
hey! is the server bugging to somebody else?
nobody wants to wait 5 minutes for image gens
@deep shoal any place to ask questions?
go for it
should be back up now
thank you!! 🙂
I mean someone already posted a youtube video about this so its understandable

working for me now, so maybe it's fixed
Welcome new members!
i am looking for an ai project to partner with on my project so who would i speak to?
amazing all fixed!
oh you are owner i see now
you looking for API access?
Who was suddenly using it so much? It was all super-recent. Looks like someone abused the system.
well that depends on collab
Hi
I think it was because of youtube video but idk
Ok, we'll start a side thread to discuss
Youtube
Hello there!
Yo!
Hello
Hi all
What happened? You got famous on YouTube, or something? Suddenly everyone wanted the thing?
two videos from different Youtubers
not huge view counts, but enough lol
added 4,000 to our discord overnight
I mean i immediately came here after that youtube video.
Leonardo <> Met4ickal
I got here first through the reddit ads 🤓
Hi
Reddit its blocked here.
I'm most followed artist on playgroundAI, it has 2x bigger server. But rly like images on your site
/imagine a blue cat eating a cheese pie
⚠️ Hey there! Please be informed that Leonardo.Ai is currently only available as a web application. /imagine and prompting for image generations do not work in this community server. Thank you for your understanding!
If you have yet to get access to the web application, kindly consider making an quick self-introduction in #👑│introductions and then check the #✨│priority-early-access channel for further details. Thank you! - The Leonardo Team
ok tag me when you got bandwidth
I just watched a video on YouTube about Leonardo and I want to give it a try
@civic violet if you haven't seen it, an invitation should be in your email inbox by now. Thank you so much for helping us out today! I've taken notice of it 🙇♀️
Welcome to our community! Do consider making an quick intro in #👑│introductions and then following the instructions in the automated reply 🙂
I did get it at exactly 4AM
Leonardo is pretty cool guy, eh generates images and doesn't afraid of anything
can you add decimal number to guidance scale? or is hard to update
I might have put in a good word (I actually didn't but let's pretend)
Welcome as well! Do consider making an quick intro in #👑│introductions and then taking a look at the automated message for more details 🙂
Not possible I believe
Hello guys
Hello there!
@marble moat would this be possible any time soon?
What is this “token” thing I’m seeing? Not sure if I like where this is going.
daily limit to how many images you can generate
I got 250 tokens
I got 200, I think the team is onto me 👀
You gonna get 10 tokens tomorrow

just came here from youtube
is there a way to "add" weight or other prompts modifier like in midjourney?
not very clear im sorry
not yet, it's being worked on!
Welcome. Please go to #👑│introductions to make an introduction and you can read the automated response by J.A.R.V.I.S for further details! 
amazing thank you!!
What's this thing about tokens I'm hearing...? BlueWillow does not have tokens. Just heard about this and already they are giving out tokens... Will it last... I guess we shall have to see.
bluewillow doesn't have tokens because they're using this first period to gather images and train their next models, just like midjourney did with their rankings. This tool already has lots of really cool and powerful solutions, they're much closer to a full product compared to bluewillow.
It's supposed to be released later but because of the large influx of the users the Team had to release it sooner.
Helloo everyone
Hello there! Please go to #👑│introductions ' to make an introduction and you can read the automated response by J.A.R.V.I.S for further details, if you haven't done that yet! 
where do you see your tokens??
@analog nebula
Why does ideate need credits now
What the hell
Hi there, nice to meet you!
Under the Generate
too many users unfortunately, we needed to limit the amount of requests in someway since our computational power is still limited (till we are able to expand more with adequate funding)
Hello there! Please go to #👑│introductions to make an introduction and you can read the automated response by J.A.R.V.I.S for further details, if you haven't done that yet! 
hi Radion, welcome to Leonardo
Aw man.
Yeah it's unfortunate but an hard decision to make
We currently are limited in terms of how much computing power we can acquire and there is a large amount of users. 🙇♀️
I mean the solution is very simple, we should all share leonardo on social media so they get all the funding in the world 👀
or
or
we culd do that
AND get people to share their gpus
and give em EVEN MORE POWER
the issue is that unlike stable diffusion, nothing guarantees that their solutions work on smaller gpus 🤔
I think it's easier for this kind of company to have a single set of GPUs that works for them, and they build and optimize everything around them
but I like the way you think 👀

man ideate breaking again
i NEED TO GENERATE THICC GOTH WOMEN
I do this on EVERY image I share on twitter (5 or 6 per days)...
LET HIM COOK!
IT NOT MAKING THE PROMPTS
The world is such a bad bad place....
know what
imma risk it
i'm NOT gonna use ideate
I dont have twitter 😢
wish me luck bois
..wait, what?
Hi, I just got my email for early access but when I try to log in, it asks for a password and I did not set up a password when I requested the early access. Which password should I use? The one for my email account?
Just implemented few hours ago
Why are we waiting that long
Try to click sign up button
time to create an account... there's a super cool community there
Yes, I did and it asks for email and password
use your email and enter the password you want
I am already whitelisted
I am still seeing the Dreamstime logo watermarked onto images, guys. Using the prompt: "cute cartoon fawn deer sticker, vector illustration, white border, white background" I just got 4 images and every single one has the watermark.
I see, could you kindly try using the sign up feature instead?
maybe you have to log in with your gmail account
Because of the large influx of new users.
this one
I know that you are whitelisted, but you still need to create an account.
if we blanket twitter + other socials (LinkedIn, YT) in coverage, we'll get our funding much quicker 😎
Ok the token thing makes sense. We’re all probably overloading the system…. But, could we get a ‘credit’ for images we don’t keep?
Was trying to create twitter few years back and i failed miserably 🤣
@spark cosmos @full atlas We did it..
Ok, perfect. That is what I needed to know. I thought the whitelisting also implied direct access to the app. Thanks!
are you using magic prompt? if so, it's a known issue, we've got a fix coming soon
No problem 🙂
It’s ok, as long as you have $44 Billion, you too can crash Twitter.
SUIIIII
I mean... the account
that's super interesting
I'll try making twitter account rq
Nope, Cute Animal Characters, Leonardo style no negatives, Hmmm... just reloading to a new version...
we'd love to, but the bottom line is the generations are what use GPU
we will be bringing in a base paid tier with more generous allowances and priority generations
let's all use this same hastag #leonardoAI cool!
Bruh....
yeah!
I had a twitter account?
you? how can we know that'
personally I've been spying him for 10 years, so I know he indeed had a twitter account
hahahahahhahahahaha vencAI
YAY! It's gone.
We are going to use #LeonardoAI in twitter right?
yes
When I signed up for early access I got an email welcoming me and telling me to join the discord. Is this a whitelist or do I get a separate email for that?
Hello there! Please go to #👑│introductions to make an introduction and you can read the automated response by J.A.R.V.I.S for further details, if you haven't done that yet! 
Js
Just did it 👍
Then you have to wait for the whitelist email to get sent.
It may take up to 24 hours due to large influx of users.
Ok thanks bro
👀
I can see priority access chanel and already fill the form , how should I know i have final access?
You'll get informed by email.
Hi I am new to Leonardo
Hello there! Please go to #👑│introductions to make an introduction and you can read the automated response by J.A.R.V.I.S for further details, if you haven't done that yet! 
🤣
Pretty sure it bugged
oh maybe just a normal queue
I managed to generate personally, maybe the servers are still a bit on fire?
true
Bottom still loading but top already loaded
Wait
Will i lose credits when my generation failed?
I hope not, worst case we can warn the team and they'll do something about it
credits are new, so I have no info
need to outpaint this one a little
nope!
yea
That's good news.
hi just got my invite but it says all images are public unless you refine ,so what does that mean
That means your generated images are going to be public, but idk about refine
so everyone can see your images
and now we get 1000 tokens and not unlimited already
How did you do that?
@spark cosmos he's hacking 😓
Been here for over a month and I got 200
yes with the new update and image is around 5 tokens
That seems weird.
i thought it was unlimited images thats what they all say on youtube
Elden Ring, Dark souls, bloodborne Unreal Engine 5, Cinematic, Color Grading, portrait Photography, Ultra-Wide Angle, Depth of Field, hyper-detailed, beautifully color-coded, insane details, intricate details, beautifully color graded, Unreal Engine, Cinematic , Color Grading, Editorial Photography , Photography, Photoshoot, Shot on 70mm lens, Depth of Field, DOF, Tilt Blur, Shutter Speed 1/1000, F/22
It was, until today.
That changed today, because of all the new people hopping on
Hello! 
Please be informed that Leonardo.Ai is currently only available as a web application. /imagine and prompting for image generations do not work in this community server. Thank you for your understanding!
By the way, currently we are having an large influx of users but please do make an quick introduction in #👑│introductions and then read the automated response from the bot for further details 
Yea
I guess some people were hammering the site with thousands and thousands of requests though
Tokens are real inconsistent if that's the case
wow hope they dont start charging or its back to playground ai and blue willow
Any way to get the same character and art style?
I mean its daily credits not monthly
They planned on monetizing for a while
You can make your own style using Dataset
how long has this app been working
They've already stated that they will announce paid tiers later on, while maintaining a free tier
There was a bug in token allocs.
Basically, normal free users get 250 a day.
but still app is really nice
It's fixed now
Ahhhhh. Alright. Fair enough
Hey how to get the bot
for me after 10 min I receive invitation email after whitelist
I just started using the app. This is absolutely amazing!
Hello there! Please go to #👑│introductions to make an introduction and you can read the automated response by J.A.R.V.I.S for further details, if you haven't done that yet! 
Should I reload? I'm still at 192 of the 200 I had at the beginning of the token launch?
the quality of image in terms of details is very bad for me , what custome model you choose?
The Tokens will be reset in 2 hours anyways
No custom models, just started with the default ones but I am getting high-quality results
failed generations won't count towards your token allowance
i been playing around you can also change the scheduler
Please dm me your platform username
Seen in some blogs already, really looking forward to this one, pls...
Btw if you hover over the allowance you'll see the reset time
It happens same time every day UTC
only thing im having issue is image enhance dont seem to work and also download hope they fix soon but all is great
Hey dear Leonardo's - I got the access email from you - but I can't log in as I did not define a pwd yet - the discord pwd is not working ...
please click that Sign Up
then you can put your email and the password you want to set
@abstract orbit I do enjoy blue willow and playground ai, too! So much easier when you don't have to worry about imposed limits
my just shows all under the generate botton how much i have when refill and what it costs for image
can you screenshot it ?
Should look like this
i just wish there was a way to keep some works private
There will be early next week, but it will be a paid feature
how long will any images be saved?
is this token thing new? i swear it wasnt there this morning when i first started using this
forever
Yes new, to help manage load
Just implemented few hours ago
Hello
ahh ok lol
Am new here
my looks just like that but when i hover over it tells me when the refill time
I watched the token come into existence as I used Leonardo today lol
in the last 3 days the community grew almost 4k new member, and thats only on discord
What if I want to have the exact character and look as the image prompt. How can I get the same character but different poses etc?
so i will always have access to my images even when i made them in the beginning? This must be pretty storage heavy...
⚠️ Hey there! Please be informed that Leonardo.Ai is currently only available as a web application. /imagine and prompting for image generations do not work in this community server. Thank you for your understanding!
If you have yet to get access to the web application, kindly consider making an quick self-introduction in 👑│introductions and then check the ✨│priority-early-access channel for further details. Thank you! - The Leonardo Team
Like, the result I got from the app was great, but how can I get the same character in different poses and etc? Like same character consistency?
hope the private thing is for all not just paid as other sites let you do private for free
Just paid. They mentioned it earlier
that sucks
There were a few people I remember had access to testing out private generation. But that was probably a month ago
Private will be key, with how often now peoples prompts and things are taken and passed around. Takes ages to refine a good prompt.
i hope they know people might not do as much work /good work because of that we should be able to share what we want as they say all work belongs to us
Well it costs server time though for them. So expecting things to be free is a ton to ask. I just hope for reasonable fees.
This tool is just lovely. A "clear" button or "X" : would do wonders on the prompt bar.
is this server in china ? lol yes that what we need clear button in prompt bar
there will be a free tier!
have fun all and a great day
Hi i am new here , I like to try this out, Can someone share me how i can do that ... Thank You
What can I do?
Hello there! Please go to #👑│introductions to make an introduction and you can read the automated response by J.A.R.V.I.S for further details, if you haven't done that yet! 
It seems like your training is niche based on gaming industry. Is that just my perception?
Are we hoping to keep prompts private for competitive commercial reasons? I get having prompts be private for people making less than savory things, but I'm a little confused here why you'd want this stuff private outside of that?
commercial reasons
There are a ton of models that are able to make gaming related assets. I've used it to make all of bestiary 1 from Pathfinder 2e
you can let the team know, they might allow you to have private generations for professional reasons
Plus theres a market to sell prompts as well. So having your work out in the open is a real problem. Plus I use images for cards, shirts, prints... Again I dont want my work out there to be picked up and reused.
Isn't AI art copyright free?
The problem isn't on the Copyright
but the prompt that was used to create the image.
So, people sell the prompts they use to instruct an AI on what images they want made?
Some people doesn't want to reveal their prompts to make the image that's why they want to make it private.
not just prompts
You can generate image to use it on prints, shirt, etc and some people doesn't want others copy their prompts.
Some yes
does upscaling an image usually take long?
Yes I do sell mine. But on top of that, it takes me hours, upon days to refine a prompt out so every time you use it, it gives you what your looking for. Having someone walk along and just pick that up is frustrating.
Large influx of the new users
It did not until the influx of people
It took me 300 seconds to generate images
oh ok, thanks
Thanks for the thought
Hello Artists
hi
Hello there! Please go to #👑│introductions to make an introduction and you can read the automated response by J.A.R.V.I.S for further details, if you haven't done that yet! 
How long does it take to get the email after submitting the form?
It may take up to 24 hours because of the large influx of new users.
But i got mine in under 3 hours
Ok thanks.
Me 4 days XD
🤣
Hi, are you a bot or a real human?
Hi, in my perspective I think I am a human.
Ok 🙂
They need a flux capacitor maybe?
Oh hey there, almost didnt see you
lolol!!
How do you unlock the Advanced Features in datasets?
There's a lot of junk in my personal feed, which would take hours to delete one image at a time. I hope at some point I can sort them into folders and remove in bulk the stuff that is really just trash from failed experiments.
coming very soon.
You can learn a lot about prompt engineering on YouTube, although I like the simplified web interface you guys have here.
Still good things you can take from "prompt engineering" videos
So how do I use this thing? The AI I mean.
#1058409094421155870 message read this
I'm refreshing my mails every 5min like a monkey with the hope to get the invitation 🥹
btw where is the Contest channel?
