#images-discussions
1 messages · Page 2 of 1
Why did the dalify bot get archived here? #🎪┃dallify-info-archived #🎨┃dallify-1a
But why? 😩
It ran over a weekend but has been removed from the server 
Ah that’s a shame, maybe it was a cost issue
Hi can dall E make an avatar with a picture of us ? Like Instagram or Facebook profil picture ?
gm I can't log in even if email and password are correct, why?
Is there like a suggestion section for DALL-E?
Is it just me or is dalle like really unuseful?
Like why is it showing me flower pictures for iphones
It's not exactly true, it does generate images even if it isn't what you requested. But I would like to see something along the lines of writing something up the AI generates it and then you can add clarification to what was generated and order the AI what kind of changes it can make to match your image. That is a serious overlook and lack on Dev's side for not adding something like that. Since I did get quite a few good art but I wanted to clarify what to be changed in what I liked and I didn't see that option
Dall-E's data cut off was some time in 2021/22 I believe so it won't have any reference for what that phone is
It might also be retuning stuff that looks like it could've been taken with an iPhone 15 Pro Max (like stuff tagged with ShotOniPhone)
see these two prompt styles. Specifiying how the phone is the subject can help
Sometimes, when a server-side problem occurs, Dal e does not work and the credit drops
I want to know how Dall•E comes up with text generation. In this case, the prompt involved a book cover with a photo of a dog on it.
https://media.discordapp.net/attachments/717493861177622590/1061015088846340247/319139033_5569370313161632_6966937516077384983_n.png
The other one I'm even more curious about since it managed to recognize the subtitle you often see on books:
https://media.discordapp.net/attachments/717493861177622590/1061014404960891030/image.png
"Ibe my poot" ||lol|| part
AI image generators, including Dall-E, aren't really capable of producing real text
basically, they know what letters look like, and that letters go together, but they're not doing any OCR/NLP stuff on source images and processing them as words and sentences
Some words that appear in the source data a lot may come out more easily than others
One time it managed to produce "poop" when I had a prompt for a stained glass window of a "poop", which I thought was an impressive coincidence.
O well, I think it is neat that it recognizes letters go together
yeah I managed to get it to spell 'wine' right on a billboard
at the very least it does know you want text
Can DALL·E remove the DALL·E remove the credits on images?
aw. And yes, I did reset my browser
okay it generated same prompt 3 times and used 3 credits ^^ the site is slow though
So I can't even access them ^^'
it seems unlagged now ^^ but in case somebody sees this it's not just you.
Yes, but it literally has "phone" in there, so it should at least make a photo of a phone instead of Flowers, like theres no flower called "phone" im pretty sure
how?
how I can play with "general art" and get images?
yeye, as I said it may draw from things that have i phone as a tag, like images taken with one
hi guys, how do I use dall-e?
I registered and tried to continue a picture, but I get an error underneath and it stops just a hair away from the end, it also scales my credits
I assume the word stalking is a violation? "3d Render of a werewolf stalking prey at night, digital art"... gota love how we censor ourselves so completely these days. please sensor me more.. i crave it.
well you need to work around certain things. Like I could not generate Heterocephalus glaber because it has "lack of things on the body" in its English name
I wish it told you which words made a prompt considered to be breaking the rules, sometimes it's more arbitrary than that and hard to tell and I had to try a few times until it stopped saying so :p
there might be (non-ai) font creators that kinda do that
New to the discord but have been comparing different AI generators for the past couple weeks. Is "Queen Elizabeth II" a banned word on Dall-E? I can't get any prompts through that name her specifically
positive vibes
create logo for memoria tours company in Dubai with purple color
create a oil painting of lord Ram
make a sun
make design
Make a hoodie with a design including the text tataki
Dall-e-sound system
How do i delete a dall e account
are you sure?
l
oi bebbes
Tip #2
Never type 'hot girls' in the description box.
Tip #2
Never type 'hot girls' in the description box.
How do i create a side view of the same character
Dalle was advertised to be a wonderful Ai image generator that could replace artists. This image isn't up to that standard and neither is dalle. Even though open ai basically advertises it as the best image generator ever, but generated that.
thats a good image
at least it tried
also it wasnt advertised to replace artists thats a bit overboard
Midhunchakkaravarthy
could someone help me with chemistry?
on... dalle chat?
nice, what room of this server?
Hey people! Is there a way to make Dall-E modify an already existing image?
It basically was. On the dalle landing page it makes Dalle seem like the ultimate artist that will generate anything you ask it to generate in full 4K ADHD. But by judging by its recent generations, it doesn't seem like ot
adhd? thats.... ok
1080p ?
oh ok
if dalle doesnt generate what you want. maybe ur asking something it isnt up to date with
also it may not generate EXACT things
kinda defeats the purpose of asking it to generate a image
"Oh, I want you to generate a picture of a tree, but if you generate a picture of a car instead, its ok"
☠️☠️
no... it generates trees ofc
here i put new york city
Really? I asked it to generate a picture of a iPhone 15 pro max and it literally generated pictures of tree leafs
Why are the buildings round ..
maybe because it wasnt up to date with it.
wdym they're good
Bro what
Buildings are squares... Im pretty sure there arent many circle buildings in new york
No bro its not accurate like its circled and the colors are mixed together
Really Good Ai!
ok ok i promise i looked up tree
Like if I want a picture of a tomato, I really want my tomato to have 2 circles instead of 1!
That looks like it just googled tree☠️☠️☠️
What
is dalle down ?? "Something went wrong. Please try again in a few minutes. If the issue persists after trying again, please contact us through our help center at help.openai.com."
Is Dalle down for anybody else?
If is working for you, can anybody give me a hand with a img edit? 😢
Yes. Cant access it for some reasons ... 😭
image generation works fine but image edit doesn't work...
nothing works for me... "Something went wrong. Please try again in a few minutes. If the issue persists after trying again, please contact us through our help center at help.openai.com."
same
is dall e down rn?
yeah
i have the same issue
me2
works again, thx
I’ll check this channel more often
Hello, I'm new to Dall-E and I have a few questions. First, is it limited to square images? Second, where else can I use this besides OpenAI.com?
Hello. Everyone!
Is it possible to create image that background is transparent with Dall-E?
Please help me.
Hi! dalle2 can't produce transparent images, but you can generate your designs on white background or any other solid color backgrounds and remove it using photo editors like photopea, photoshop, canva, bg removers, etc
Oh, yes. Umm, it seems similar.
How do they make that?
sorry, not really sure about it
oh, my god.
Any solution?
it's not really dalle2 or openai related, so I am not sure how can we help you.
😢
Can we use the images as a fictional story as a tweet?
Hi, sir
About the project; if someday the software has cameras in natural areas located all over the world, things will get out of hand when people ask chatGPC to create a documentary about the subject called Forests and wetlands of Malaysia...
Hello there, this channel is for discussions about dalle2, all other ai discussions can be held in #ai-discussions
Asked dall-e to create this for my website, looks amazing
football player winning world cup
create banner for website
2 male wich and 1 girl warrior girl pumped muscle one wich have blue mohawk and second have long hair on forest portal opened on background - @maiden hedge (fast) - @maiden hedge (fast)
galatasaray photos
hunter carrying his catch
hi
Does anyone have a suggestion for editing a selfie into a professional headshot for linkedin? 😅
Guys I was trying to design a logo in DALL E and how many ever times I try, it always shows the wrong Spelling on the logo. There was one time when i told the model to print the word ANALYZE, it had printed "anamilae" or something similar. Is this a problem with everyone's model or has the model not learnt to recognize and print words yet?
AI image generators currently aren't capable of accurately producing text
Does anyone know anything about igem
/love peace instrument wood
looks like AI art is being accused of just scrap-booking existing (often copyrighted) images - companies like Stable diffusion being sued by big guns like Getty Images. Any evidence that dall-e does more than combining bits of existing images? Seems they are going to have to learn how to make art from scratch
create image for cancer awarenes
pickachoo
the generators don't actually work by 'combining' any images. The model is trained to associate words with a certain 'look', but the generations come from a random noise pattern it iteratively improves until it reaches a confidence that it matches the prompt words
I'm not entirely certain about where OpenAI got their training data but afaik they did a lot of cleaning and that likely involves removing clearly trademarked stuff
like, you never really see watermarks in Dall-e images
though it'd be funny if Getty is suing given they're notorious for stealing images and slapping their watermark on
Getty is suing Stable Diffusion because their watermark literally appears on some generated images, just in a warped fashion. Thus the 'noise' seems too subtle there. Quite how different other generators like dall-e are I am not sure...
(not watermark - their label)
yeah in the case of SD i can imagine the training data just isn't being cleaned by people making the models. It uses LAION 5B usually which is uncleaned but still has tags, so I imagine, due to how Getty plasters watermarks on everything, the model will associate the tagged words with that watermark
yeahhh it's still a watermark
tbh this should just demonstrate how awful Getty is
OK!
basically I think dall-e is more insulated from these issues because of the data cleaning OpenAI did
and that's sort of what makes dalle generations stand out, even though the method is pretty similar to SD iirc
cleaning will help, may be questions over just how close images come to some of today's artist's work, but we will see with upcoming cases
yeah. Thing is, I can imagine that the issues of people selling work with similarities will fall under the same legality as counterfeits already do
especially if you're trying to pass it off as something made by that person
in terms of claiming ownership yourself of something generated in another person's style? well, the law's already super fuzzy on that with handmade stuff
Where can i find chatgpt for germany
Hey guys, im doing a quick research on generative ai. I was wondering, how many users use dall-e on a monthly or daily basis?
I think they announced roughly how many weekly when the API was announced in September (maybe). I recommend looking for that
Asking for 46, always giving me 44. 😄
How to create art with chat gpt?
you cant'
I spent all the credits issued for a month for generations in 1 and a half days(
Hey guys, images generated by dall-e with the free 50 credits is mine ?
Where is dall e bot? I need to pad50 something
you have full rights to use them however you want
The dallebot is no longer around sadly
😦 what the heck
how to do videos with Dall e
how to make a text to image discord server like mid journey and bluewillo ❓
I’ve heard it usually messes up on certain texts and stuff 🤷♀️
Does a dall e art image need 5 stars to get in the hall of fame?
They need 10
Oh
Is this good?
This channel is for displaying art. Reply to an image to discuss it, or move discussion to 💬┃dall-e-discussions. If you want to generate art, visit DE2 at https://labs.openai.com/
Very interesting work! This post is a little problematic here because the server prohibits promotional posts. Also, the writeup does not seem to discuss how DE2 was used as an inspiration as part of the creative process. You may not have been aware of our server policies on promotion. Perhaps you could revise your posting to make it more clear how it is relevant to DE2 as a design tool? We may need to remove the posts unless we can bring them into better conformance with server policies on self-promotion.
For guides to tested styles please see #📚┃resources.
Hello, I'm asking Dalle to add text, but somehow it seems he translates it. Is it possible to tell it to write the exact words I give ?
Success
DE2 is not designed to work with text at all.
DALL·E 2 is not engineered to properly generate text. While it may get a few simple words right now and then, that is the exception rather than the rule. If you need to add text to your image, consider using an external image-editing software.
Nice picture, but in the future please use the gallery channels to post your DE2 images. This channel is for discussion of the model.
Hi chat
create for me a picture of a nursing technician doing her work with love, disney image
did dall e get worse recently
No, the model has not had any recent changes.
oh
But one thing is that IMO a lot of DALL-E generations have become worse in quality
Did the parameters or database get lower/smaller?
can this bot be quiet
The perception that quality changes over time is not uncommon, but difficult to replicate due to the uncontrolled nature of the prompt testing people use. As far as we know, the actual model has not changed since it was released, but it is possible that some pre- or post-processing steps have been modified. We have no exact documentation on that. Personally, I see no difference in behavior.
Got it
Especially considering how much DALL-E 2 varies in quality
Also, will a successor to DALL-E 2 ever be released?
Or do the devs have no interest at the moment?
We have no information on the development plans inside OAI.
Understood.
I'm planning to have a hangout with around 8 people where we all use our 50 credits to create dalle generated photos together.
I'm guessing there is only a 50 credit limit per month per person limit (and they enforce that with 2 factor), correct? Or is there another way they enforce, ie a 50 credits per wifi network or server limit?
does anyone know how to get an image masking tool where you can use a brush stroke to get a transparent version of an image? like in the dalle playground how i can add transparent image
nice
hello, I am wondering if anybody wants to help me generate a logo - I know that there are many projects outside of midjourney for image generation, but I am curious if anyone wants to play around with some prompts to generate a silhoutte style image of a human brain and artificial intelligence type half brain half mech, I am working a project braiain.com and if someone wants to take this on, I would love to use the finished product and then follow up with a article on the site detailing your processes and of course if you are working on a project we can feature that as well.
How can i upload an image and generate images "like" anything inside that image ?
Hello, I totally understand and sorry for the confusion! Here I would like to share a screenshot from the modus-operandi on how DE2 was used creating the installations. I actually wrote to Natalie too proposing to write a full documentation on how DE2 was utilised and in what what ways it was inspiring.
As the PDF up there is mainly for gallery spaces, and even though the credit is there, it doesn't specifically focus on the DE2 as a design-tool - but more of the concept. I might remove the .PDF from the post if you'd like? The very very initial inspiration for this installation was the fictional museum like images DE2 generated.
Thanks for responding! This is quite fascinating, and a good example of a creative workflow involving DE2. I suggest that you remove the PDF from your previous post, as it conflicts with our promotional policy, but leave the screenshot here as it is more directly related. We can talk further about your idea for documentation on this type of workflow.
That's so cool
Thanks very much, very flattering to hear that! And done so, apologies as I couldn't spot the distinction 🥹 Also please feel free to reach out in terms of documentation! I am very inspired to work on a write-up, documenting the making-of [3D modelling, coating, designing the pieces by combining various parts from various generations and even the prompts themselves] but also of how these even managed to create a feedback loop of influencing the sound-design of the musical counterpart of the project. Best wishes! ✨
Thanks very much 👼 💐
Damn bro I feel for you
hi all, can someone please explain how can I try the drawing AI myself? I remember there was a discord channel where the bot used to gave 4 results and then you could upscale/get variations for them
you're thinking of Midjourney, which is not made by OpenAI
Dall-E is acccessed via https://labs.openai.com
hello! How can I keep a style for 10 different pictures?
something like a children book with different ilustrations but the same style
Guys is it possible to upload multiple images as a feed for generator or u can upload only 1 ? Its my first time use
Can dall-e make infographics?
How can I create a wallpaper of Zoro character in one Piece
I also do not know bro
visit https://labs.openai.com/ to generate images or click the bot's "Generate image" link
The systems pretty awsome
Hey guys, im having issues with Dalle 2. When I put an image , in my case was an image of a castle than I tell it to generate additional items based on it. It's not giving me any changes at all, any tips/tricks ?
Thank you
can anyone use art that’s published under daily theme?? or there’s a way to credit AI generated art…?
Hello please a went an example to drawing with dall e
Ive noticed that DALLE generations have had a significant drop in quality over the last few months. Anyone know what’s going on?
I wrote "Raiden Shogun from Genshin Impact" on the DALL-E Generation, and I got completely different character that is not from genshin impact
Anyway, is there is a way to make the Image generated to have better quality?
It says I can't post links
I got it open 👍
are we allowed to turn stuff dall-e makes into nfts and profit off them/use dall-e commercially?
Big brain time
I want to make a picture for Lionel Messi in Real Madrid shirt?
You can generate it in Dall-e
Guys how much does dall e cost
you get 15 free credits a month (they reset each month), and it's currently $15 for 115
Hello all, that's a pity that chatGPT does not mention DALL-E when you ask for websites that can generate images from text 
You can use the OpenAI API and make a bot yourself using node.js or python https://beta.openai.com/docs/guides/images
An API for accessing new AI models developed by OpenAI
am I allowed to mention that you can use OpenAI API to generate images at $0.020 / image (1024x1024 resolution) and also use Bing Images if you're in a supported region? 😅
Os sci fi good?
Can’t we not generate image in discord like the mid journey ?
Question folks, can I provide example image and DALL-E generate similar?
Does anyone know if there will payment methods other than Credit card? I would love to use paypal
anyone knows if any AI is generating images with vectors ??? That would be usefull for me as a designer
Can I generate pictures in discord like in Midjourney server?
Ok
Do I have to pay to use dall-e or chatgpt?
is there a token for openai ?
What is "wrong" about this request? Since when was "Cringe" illegal on Dall-E?
what is dall e
How do credits work? do you loose one per request? or per download?
anyone using the generated images to print? offset or laser? quality ???
What is it? Is it mean that I have to pay to use DALL•E?
Eventually, yes. You currently have 19 generation requests of credits remaining.
Per request.
You can ask ChatGPT to generate SVG code for images. It's pretty simplistic results though. It hasn't been trained to produce images.
Does anyone know some cool style prompts?
can i delete a image i uploaded? i uploaded my self but now i wanted to delete it. or should i not worry and they are private?
how do i upload a image
hi, i am trying to use dalle 2 to remake my discord profile pic which was made with dalle mini. when i enter the promt and press generate it says "this request may not follow our content policy." this is my promt "A cute plastic duck figurine with a yellow body, orange beak, and green head inside a clear snow globe with snowflakes swirling around it" can someone tell me what is wrong with it so i can change it?
So if I upload and image can I run a prompt through the image ? Say if I upload an image of my dog, could it paint it in watercolor?
Dall-E does not have the type of style transfer function you are describing.
We currently don't support transforming the style of an image into another style. However, you can edit parts of a generated image and recreate them in a style you define in the prompt.
Attention Off-topic chats have been cleared by a moderator.
Contact support@openai.com about deleting your history
Hey guys am trying to generate how a ranch/farm will look like in the future anyone has any suggestions for prompts?
I generally explain in 5 or less words my core concept, then a comma separated list of attributes.
"A futuristic farm ranch, sci-fi, digital art, vivid colors" for example. idk if that's good but it's what I would try first
Superhelpful thank you
How do I have a logo created? Any help appreciated
Hello guys
test
Hello !
Quick question - Is De2 able to edit and uploaded image? - Example I upload a phot and tell to transform it into a drawing based on a specific style.
(If not, any other tool recommendations?)
Best,
B
whats dall-e pls fam?
Is there another free way?
Does anyone have an idea how to prompt text/letters correctly in Midjourney or Dalle-2? It always gives some random letters and ignores my input ..
great!👍
maybe you could try the Python codes?
nice
.
Why does my Anime AI art misses Line-art?
Like, it always looks blurry for some reason, like it is drawn on a paper with pencil without any line-art.
This is what I use "Anime key visual of an anime girl; official media; by Hololive; white hair; purple eyes; purple lightning; Thunder; wears black slacker suit; wields a katana; poster"
Can you answer a question?
description of a double sided tape
how can we have credit without paying credit?
hi, can you explain in details what is placemaking?
why i cant enter the chatgpt ?
tnaks! i will look into it
cat
Hi all, do you know if with Dall-e I can create an image of a specific HEX (hexadecimal) color, telling the machine to use for example #FF0000 (red color)?
Or maybe use other color codes like RGB, CYMK etc etc?
(I couldn't find anything on the web)
sorry I forgot to say Happy Chinese new year 22/01/2023
I think the problem is with mentioning a specific personality, in this case King Harald V which isn't allowed on DALL-E
is there a way with an initial prompt to get dall-e 2 to zoom out more, show a more midrange or far shot? i.e. trying to generate a rink full of hockey players
Try to play around with "photo depth"
try refreshing the page a few times and it should let you in
thanks - like 'medium photo depth'? or what terms should i be using.
I am not sure, but as you described the problem... The answer is the depth of photo should be... maybe even "scale out photo to fit more people" I'm new to this stuff, just tryin to help 😄
why cant i access chatgbt
Is There a Chat premium?
Wow 150 credits are kind of expensive right?
if any mods are in here could you message me please
help? how getting started?
Thank you for the answer !
spam refresh button until it lets you in, works for me
i had it for a secound
Help pls
I want lot of information about dell-e
Someone got this from "Dall E" but is they a way to se what the person type to get this?
Hi how can i save my "art" generated like this for example:
you see, there's the name of the guy on top of it
hi folks
so i encountered 2 issues w/ my dall-e acct.
- i tried to reset my password - have yet to receive the password reset email
- keeps telling me that my phone # is already being used when im asked to verify ( well duh, i set up my acct with it)
anyone else had this issue?
is dalle just art
Who’s your favourite artist for dalle? Mines probably Michael Whelan or Ralph McQuarrie.
ayo anyone know the site which is like a database for a bunch of modjourney/ dall e images
I've been enjoying Ivan Aivazovsky lately.
morning
It depends on what I'm trying to make. I love Frank Frazetta for fantasy stuff (especially dark fantasies). Like Seamus said, Ivan Aivazovsky is great (especially for boats). Albert Bierstadt is great with scenery, and adding Thomas Kinkade with him is chef's kiss. Recently, I've been trying a little more of Goya which has had some neat outputs. There are some others that I use, but they're living artists, so I won't say their names here 🙈
Is there a model that can generate text, typefaces, or images with text?
By text I mean stylized images of text not just the text like for signage.
Like “sign to take your shoes off at the door” that would generate stylized text of that meaning optionally with an accompanying image.
Or even just generate the words.
I don't know of any image generation models that specialize in words/signs currently
Hi everyone, new here and I'm enjoying creating images using Dall-E!
Was just wondering are the generated images safe to print publicly/commercial use or is that even allowed (based on what I read in the terms of use it should be alright, but just wanted to make sure. lol) ? Was planning to give some to my friends.
Does anyone know how do I instruct Dall-E to color a certain part of a line art, for shading and coloring purpose, without morphing my original lineart into a completely different picture?
yes you have full rights to your generations
Wow that's awesome. Thanks man
hi! im new with dall -e and i want know if it posible get a custom image for a niche like "anime" or similar uploading a normal photo of a person. Im looking some info but not find any instructions to get good results.
Put a character or author's name in your prompt for more accurate reference
I'm trying to upload a selfie photo and then put some instructions on it and have the AI transform it into something more artistic/different but it does absolutely nothing after writing instructions and hitting generate, I don't know if I'm doing something wrong but... it literally doesn't change me nothing
Style transfer isn't supported by Dall-E 2 😦
ooh, so only can make a "random" art right?
It can only generate, it can't modify. You can erase part of am image and generate a section of an image, but I can't modify/alter a preexisting image
Is it me or did dalle 2 have a huge downgrade?
hey guy's how do you generate the same image with a prompt? i already have an image generated but i want that someone types a specific prompt to get the exact same image as mine. how do you do that? i remember it has something to do with adding something to the end of the text like -sdt2345
But i have forgotten
thanks
just type in dalle 2 in google lmfao
Bros, you can't generate stuff here. this is a channel for Discussion about dalle2. If you want to generate images go to : https://openai.com/dall-e-2/
and click "log in" or "register"
I am loving this app but it won't let me add credits. I updated my payment on the openai site but I can't seem to get the buy credits page to show up or load after I tap the buy credits button
I assume you mean trying to buy credits on https://labs.open.com
Hey, pls remember to try to be respectful towards other users even if what they're asking is misplaced in the wrong channel. Especially important when not everyone has perfect English skills 🙂
Anyone get a good prompt for human shaped animals
Is there any Help page that goes over all the different classifications which are possible, as well as how Credits are broken down and used? When I click "help" the help pages simply don't load and freeze up on the website.
How to use dall-e?
btw your link is missing ai after open 
totally didnt click that link cause i was lazy to type the link to dalle
Hello, I'm having a hard time creating a black male character with Dall-e. Could it be that the artists I use "by" don't have any black characters drawn? I also tried using actors for reference and it does a little better, got best results with John Boyega. Anyone know why that is?
why did dall e added credits bruh
" DALL·E API is billed separately from labs.openai.com. Credits granted/purchased on labs.openai.com do not apply to DALL·E API. For the latest information on DALL·E API pricing, please see our pricing page."
Is all I see and can find I go through the openai.com page and it doesn't show a set pricing just by the image pricibg so I clearly don't know what I'm doing now
Hi! is there a way to upload an image and edit it so it has a chosen style? without erasing anything, just the same image but as if it was painted by Monet or Picasso
where do I ask for help with nodejs api? 🤔
Gotta be honest, Dall-E has completely fallen off in quality to a degree that I can't see myself using it anymore unless some radical adjustments + improvements are made to the standard image generation model.
When I started using Dall-E in August 2022 for my art project (creating highly specific imagined images that I could then paint in real life), Dall-E was remarkably good at illustrating very abstract yet exacting concepts. You could dream up just about anything and get beautiful results, especially for painting styles.
Now it is absolute garbage by comparison. Even simple concepts, like one person doing something or making a certain face - always come out garbled and unrealistic. It no longer gets faces, groups of people, or hands right at all. At least not with the styles I'm trying for (medieval art), which could once be generated with these things no problem.
I'm not going to state specifically which services I'm referring to, but there are several other image generation services out there creating stuff 8,000x better than Dall-E with higher resolution and even better pricing. If Dall-E's developers want their model to be viable for working artists again, I'd hope that some work could be done to bring it to its former glory. It used to be so great. I'd love to see it again
DE2 does not have the ability to color images in the way you are requesting.
Attention Off-topic chats have been cleared by a moderator.
Thank you, for artist like me it would be a great feature to save weeks of working, something like coloring within a marked circumference, or automatically applying shade
The functions you describe are more typical of a general art editing software. Dall-E is engineered to produce images from text prompts, with an inherently random quality to the image generation process. It has an editor, but the editing functions are designed primarily to extend or replace portions of an image by generating new content. For an overview of the typical use of the editor see the Prompt Tip for that: https://discord.com/channels/974519864045756446/1057394170777587794
Hey how do I get access to Dall-E
I have an openai account but it doesnt work for Dall-e
Can anyone help
Attention Off-topic chats have been cleared by a moderator.
visit https://labs.openai.com/ to generate images
Hello everyone, does anyone know how do you upscale , enchance quality and remove watermark from images, thanks!
Guys, but can i write on dall e in Italian or is better in English?
Sorry for the bad english lol
How many requests can we make from Dall E
How to use dall e guys 😭
Dall e. Is that a app? . Whats primary to know and some tips to create?
visit https://labs.openai.com/ to generate images
there is not an app for Dall-E 2, but there are some great guides in #📚┃resources ! 🙂
is dalle same login as chatgpt?
Is there a way to set it progressively change with more input like the demo video I was watching?
Tell me, is this site related to this service or is it phishing? beta.openai.com
It's not just you. I spent 18 credits in the last 24 hours, all of it garbage. I gave so many precise instructions, oil painting, citing specific art styles, specific painters
TBH I think I deserve a refund
yes, they use the same login information
that site takes you to GPT-3 which is a set of Large Language models by OpenAI. It's real 🙂
https://beta.openai.com/playground
An API for accessing new AI models developed by OpenAI
I'm sorry that you were disappointed in the output images. We have a few resources in #📚┃resources , #1021130377026351105 , and #❌┃tips-n-tricks that may help you
Hello, what do you think about people who post Dall E images on Instagram ?
Me pero I use openai to have more visibility on Instagram
Hi can anybody tell me if I can use a photo generated by Dall e 2 for commercial purpose ( Digital CD Cover). Or who is the owner of the photo rights?
Hi! I keep getting the same issue where any prompt I enter, the image comes back with an extremely close/zoomed in perspective. For example I'm trying to create an image of a couch, but they keep coming back with the sides of the couch cropped off. I've tried adding "extreme wide shot" in front of the prompt but it doesn't do anything. Any ideas?
Also the site looks like this for me. I've disabled any VPNs. Using Safari 15.6.1 on macOS 10.15.7
whos got the coolest background i could use for my space game
Hi everyone! Is there a way to download all the images I have made? Or auto take the watermark off? Thank you !
Hello, is dall-e capable of rendering 3D objects for permutation based design solutions, complex mechanical problems or for example so invent new key-look antibodies for medicine purposes?
Is it possible for Dall-E to ever get integrated as a discord bot?
No, DE2 does not have those abilities.
For discussion of other AI models please use #ai-discussions. Other models may be discussed only in that channel.
Discussions of non-OpenAI models are only permitted in the #ai-discussions channel.
As a community, we value dialog about the future of AI, including alternative technologies and their associated ethical issues.
Attention Off-topic chats have been cleared by a moderator.
Is it just me, but the quality of Dall-E 2 is a bit lower than before
I asked Dalle2 to create an image of a bottle of IRONBEER, my family business in Cuba 1917-1960. However, it was unable to create an accurate rendition.
how do you get it to create you an image?
you need to got to https://labs.openai.com
Thx
I am trying to generate an image of real elephant standing on a football
I am trying to generate an image of real elephant standing on a football but I am unable to get what i want through Dell.E
consider using a chromium based browser, it might help
Attention Off-topic chats have been cleared by a moderator.
I can not log in in the dall e site although I have a openai account
Do we have basically no control over what dalle generates? no control over any parameters? how do i edit one of my pics?
Is it possible to run dall-e locally?
damn. I don't know if asking this is against server rules but are there any ai image generators that can be run locally? I know about stable diffusion but I don't have an nvidia card :(
Took a quick glance at rules again. doesn't seem like asking about other ai generators is against it okay
I’m saying I would wait for open Ai to release a local download before sourcing it from git hib
oh ok thanks. how long would you say will it take?
Maybe a few more months. Can’t say precisely. Idk . I haven’t heard of a stand alone client yet for this but no doubt will happen this year.
I sure hope so. I love AI image generation but I'm also broke
wow i just entered openAI what an achievement
No, it is not possible to run DE2 locally.
Can we implemeny dall-e in a discord server for free
Where can i use dall-e?
If you get an image back that you feel is inappropriate, please use the in-app reporting features, not Discord.
just why
Is there way to change the size or length of a image in dall-e to make it fit a PC wallpaper
Hi, do you know what's the style art pls?
is one credit for one picture? 🙂 i dont want to waste the first free 50 so fast
one wish is one credit
thanks 😄
by every wish you get 4 pictures
how can i remove this when i create an picture ? the corner with yellow/blue/green cube..
I would also like to know the answer to this question
what should i do to generate a good anime character
I run out of credits asking dall-e to draw a cute fox walking in the moon, because the foxes weren't cute, and weren't walking in the moon. How should i have made the prompt?
Waiting a month. Every month it gives you few free creddits
somebody know how to connect DALL-E API to a website ?
And somebody know if DALL-E Work with shopify ?
Is it possible to use these Dall-e tools for photo upscaling?
Does any one knows how can I generate a set of images one by one based on image description, but keep the same characters ?
The weirdest thing just happened, look how every single image has some sort of (what I'm assuming) watermark
but look at the 3rd one, that is straight up legible and perfect text
Photgraphy is not that far off
yea exactly
That font is perfect
and the rectangle
how the hell did it to that without any prompting
this is next level watermarking
It is very difficult to recreate non-DE2 art. For help with identifying art styles, consider looking at the style directories listed in #📚┃resources to see if you can find a tested style that is close to what you want. For more details on art style identification, please read the Prompt Tip on that subject: https://discord.com/channels/974519864045756446/1032836287851413586
For anime character guides, please see the guides listed in #resources, or check the #anime-manga gallery on this server.
Please direct API questions to the #dev-chat channel.
No, DE2 is not designed for photo upscaling.
how to tell artificial intelligence to design an image please
thank you for developing me a code of a website with the possibility of registering members
You are posting these two questions in the support channel for the Dall-E art product. For getting started with image generation see the guides listed in #📚┃resources. Your other question may related to the ChatGPT system, which is in #chatgpt-discussions. Please direct future questions to the channel specific to the product you are asking about.
Is dall e able not allowed to replicate characters?
Like I'm trying to create pictures of anime characters fighting but its not making them
Is this for copyright reasons
Currently there is no way to create "an asset" in DALLE and store that asset (character or object) for ongoing use in multiple images, like an asset in a game development system. Even if you specify many details of the appearance, there are some random aspects to the generation process. This results in variations that are very difficult to control.
The AI is limited to images it has in its training data, which has been selected based in part on copyright availability, yes. There may be other factors that limit the ability to reproduce a character based on other words present in a prompt.
Has anyone used Dall-E to generate visualizations for use in technical presentations? For example, if I'm creating a PPTX on neural networks, is Dall-E a good tool to use to generate a graphic image in .jpg, png, or gif format of an example NN?
You probably can get general images, but it will not be able to produce anything specific to an actual network. Think of it more like an alternative for clip art type things. If there is a specific prompt you are trying to refine for that purpose, you could ask about it in #❌┃prompt-help.
Ok. Do you know of a tool that is capable of doing what I described?
Discussion of other AI tools is permitted only in #ai-discussions. You may also want to use a regular digital art software.
Discussions of non-OpenAI models are only permitted in the #ai-discussions channel.
As a community, we value dialog about the future of AI, including alternative technologies and their associated ethical issues.
Ok. Thanks much.
Sounds like you are asking about ChatGPT? Please use #chatgpt-discussions for that model. This is for the Dall-E art model.
#chatgpt-discussions Please tell me how to remove the limit on the number of characters. Now I have a maximum of 700, and the text breaks
If you creatively modify images that were initially produced with DALL-E 2, would it still be public property or would you have the copyright to that image?
Subject to the Content Policy and Terms, you own the images you create with DALL·E, including the right to reprint, sell, and merchandise – regardless of whether an image was generated through a free or paid credit.
Do you exclusively own it, or is it like Midjourney where all public people can use or modify the images you create?
Also, about modifying it?
Yes, I was wondering if it was similar to Midjourney as even they say you "own" the images, but then they further go on to say that all images and prompts produced by the bot can be publicly reused and modified (including all images made by Midjourney that is publicly shared). Is it different in DALL-E?
You should review the content policy and terms
But no, I don't believe your images are public unless you make them public
Yes once they are public and found out to be created by AI is when I want to know who own the copyright or not
Please, all the answers you want are in the terms. I'd rather not have to search it and regurgitate what it says
how to develop image on imagine any body help
I recommend reading the automatic bot message
owl
Which is better at generating image dall.e or dall.e 2
this has def been brought up before, but the quality of dall-e 2 has gone significantly down
hi, how can I generate an Image?
John Cena vs. Spider-Man
hot cat
irfan can kahveci ironman
icebreaker
bmw
white wolf with red eyes image in 8k
devil with dragon
devil with dragon
Can you create a custom logo for the name skrayius?
Lol
Hey guys 😊 how are you all
Not bad! But it’s hard to describe what we want it to do. It takes time to perfect the description so that it gives us good results.
make me a t-shirt design with mustang and american flag background
Hi there, any ideas on prompting for logo generator? It's quite hard, I had some samples using stabledifusion but steal not exactly what I want......
is dalle down?
yeah i think so
makes sense was getting so annoyed haha
Yo guys is server still down? I got some errors
Someone do a minecraft character lmfao
iraq
Id do it if i had credits
That's why I enter to here. Good approach to take users in discord lmao
Why is Dalle 2 really slow?
Normally it is not. But now the server is totally down;
I'm still getting images I'm just getting a 90% error rate
Hello all! I'm unsure if this is the correct channel for this but I was basically trying to make a pattern style backdrop with a flat design (think material UI) and gavee Dall-E 2 this exact prompt landscape of mountains as a backdrop using material ui. It just kept loading for quite a bit of time and then just cancelled my request. Is it confused, did I break it, or is this something that's not allowed?
This keeps happening to me as well, came here hoping for help
yes.... why is that?
Okii, so it might be a server down situation rather than my query being a bit too dfficult for Dall-E
i hear the servers are pretty overwhelmed lately yeah
Who are the people overwhelming the server?
👀👀👀
all of us technically xD
guess all we can do is wait
how to remove history items or images u want
flat logo design combining letter “SA”, simplistic, company logo, black and white color scheme
How do I acces Dalle 2, is there a server on discord or do I have to login to the site of openai?
Why do people type prompts in the server expecting that its going to generate for them. When it says it doesn't support image generation functionality infront of their face?
https://labs.openai.com/ this one friend
looks like openai has put all its resources in chatgpt, dalle2 is got very slow now
Makes absolute sense, it is a free API after all 😛 and their paid API is like fractions of a cent per request
I know how to solve this problem. Make dalle opensource
bot needs to learn to differentiate b/w prompt and chat messages
@silk sandal alright, I guess there is no discord server then?
I mean I guess it could become. Still the training resources will not be 😄
and if you have the resources, knowledge, and time to train the open source model you probably have already made it and wouldn't be here :3
I'm training a Dalle 2 model now
The discord server is for discussion etc, not to actually make prompts and generate images
i think now all laid off employees are gonna apply to openai
drop us the link when it finishes, maybe we can make use of that then 👀👀
I'm still making the giant database as I need thousands of images for it to be remotely good
exactly my point 😛
like literally, it has been one hour now and the dalle2 is not able to generate a single image
lazy Sunday perhaps
i need elon musk's phone number, gonna call him 😠
Are the administrators going to mention this?
I'm a little weary that it's down for this long without any communication
Hi there
You can't
Most likely OpenAI will still keep it in your account in the bg, but you won't see it
So what I'm seeing is there isn't a way to remove just one image? You have to remove all of them?
What does Dall-E?
do?
Mhm
Creates unique images from text descriptions
believe so
to remove images from your generation history, you'll need to reach out to support@openai.com
berry shady
hello!
My message vanished for some reason. But is there anyone here that has an idea when Dalle is going to work.
While I don't have a timeline for it, I have raised the issue to the OpenAI team ❤️
You absolute legend
how do you fix error 403?
https://status.openai.com/ shows that the api is currently being "investigated"
Welcome to OpenAI's home for real-time and historical data on system performance.
You sign in
i am signed in
403 = forbidden
while I has been testing some integrations, my API calls usually takes 2-3 retries before I get an image back. In other words, 7 minutes before the picture shows up on my end
yeah it says that when i do ctrl shift i, im signed in, not using vpn
(recently)
it works on my phone
hmmm, sign out, perhaps clear openai cookies and sign back in?
alright, gonna try that
I'd wait for a bit as the api seems to be having a partial (or complete) outage
yup, it was just weird while I was coding the past two hours, and suddenly it just stopped working... I had to use an hour to investigate if I did something wrong the past hour 😂 until I checked this forum
I'm french and I have a question about openAI. How do we do to make an essay because I try to do this, but he also write 20 answers... Thanks !
Sorry for my english
answers = sentences
that would be for ChatGPT, Dall-E is for images
I put in a vert simple prompt and still am getting the "request cancelled, try again?" error. What am I missing?
the api is down
Thank you! That would be my timing. haha
please, help me
@dim cradle check your DM
Hey there, is the website currently down? at least iam unable to generate images for a while now
How long take dall e to generate a picture?
It's not working right now
ho lol just when i went try it
Where can i find this info?
either maintenance or server outage
Welcome to OpenAI's home for real-time and historical data on system performance.
.
ok dac
is there like some cheated prompts that we can usee to have cool art
what do you mean
there's no "cheating" the prompts
I think that depends on what kind of art you like. You can always browse the DE2 channels, find something you like, and ask what the prompt is
Yes it's down
thanks
I wrote the same generation twice because of it, but it appears that dalle-e is still generating.
Will it waste 2 credits?
If it does, just accept it or contact support?
Hey is Dall e working rn ? Seems unable to generate any request
perfect timing 🙂 created an account a few minutes ago and was wondering if my test prompts are too complicated and registered here in Discord after I requested a picture of a puppy. Any puppy....
is dalle really slow for anyone else too
Everyone is having issues
not working for me
At least it didn't seem to eat my credit
Also, why is the OpenAI bot spamming us? Is someone trying to gen images here? Or does it always do that?
hi, can you DM @shadow bane with an image of what @silk heath sent please?
@oblique widget‘Astronaut Jane Austen, staring into the cosmos, floating through deep space, beautiful galaxies in background.’ There's an example of a prompt straight from a guide in #📚┃resources . I recommend you look through them as they're really helpful
Thankyou 💙
Its normal ?
I'm not able to generate in DallE2 either. Have paid account. Haven't been able for last hour or so. Stops at 99% give or take.
Dall E doesnt work bruh
"the request was cancelled"
ima ask chat gpt why its dall e its not working
Same here.
my designs keep loading to the end and not finishing its happened like 5 times, anyone know how to fix?
I realized how rude this sounds sorry for that. How many times do we have to say this?
https://status.openai.com
#announcements
above
Welcome to OpenAI's home for real-time and historical data on system performance.
The image generation is taking so long even for simple prompts
Yep, DALL-E website seems to be down
Yeah it just gets stuck at 99% for me , guess il try later
it works
The website still seems to not work, at least for me. (I get the message about too many redirects when I try to just visit the site at all)
yeah same for me
still seems to be down :/
Dall-e site still down?
when do credits refresh?
Just a question, can this AI generate accurate or alternative history maps?
Discussions of non-OpenAI models are only permitted in the #ai-discussions channel.
As a community, we value dialog about the future of AI, including alternative technologies and their associated ethical issues.
Attention Off-topic chats have been cleared by a moderator.
No, DE2 is not designed for that purpose.
Why does DallE send cartoons lately ?
Any idea why the bot keeps auto-replying to me then deleting my message? Happened twice in a row in the 3D renders channel when I was complimenting someone's post.
it's telling you you need to actually reply to their post (like this message)
Aha, that's odd. I don't see many direct message replies. But thanks for the info LordJord! I'll keep this in mind.
it does also reply to you in the channel but that gets deleted shortly after so it doesn't fill up
Anyone else getting ERR_TOO_MANY_REDIRECTS when trying to access dall-e-2? I have no problem accessing ChatGpt, I Have tried both firefox and chrome as well as clearing cookies to no avail.
hello,the link keeps throwing errors :/
How do I generate images help!
Hey guys can I get a help from someone?
hey guys is there anything like a dalle flatrate i can buy
dall-e 2 is free?
You get 15 free credits that reset each month, but you have to buy any more than that.
I'm wondering about a Dall-e and GPT merger. If Dall-e had the knowledge base of GPT then it could actually render scientific imagery. I wonder if that's even possible.
Hi all, thanks for developing this fantastic tool!
While using this tool, I had to adjust the text description and generate the images multiple times before having the one I liked, so I wasted many credits. When my credits ran over, I felt like I'd be gambling by buying additional credits since I may not get any image I like by the end of it.
Do you folks consider paywalling the download step instead of the one that generated the image? You could even put watermarks on the generated ones to keep people from print screening.
is it offline? there is an error when i try to open the website
does anyone know how to fix the "openai.com redirected you too many times." thing?
or is that i dont have acces to dall e 2
but im sure I've used it before
What can i do if the site of dall-e says:"too many redirects, you can't open the Page"?
DALLE2 --> Tips and Tricks - Hall of Fame (best tips and tricks rewarded with bulbs) --> so how do I find the best tips?
Dalle2 ia not working
we have to wait now ?
Dude charge ur phone
You can always sell your liver
Hey! Been testing the chat for a couple days now, and I think it works faster in Anglish, instead of Portuguese (native language). Did you also notice that when writing prompts in languages other than English?
Oh, I think I'm in the wrong channel! Silly me
Hey, everytime i want to open Dall-e i get the Error too many redirects, what can i do to fix the Problem?
yes it's down
dall e 2 made this so i converted it into minecraft texture pack
Down?
Hey Dall E folks
Is there any hint that an update to dalle is coming?
you should ask in #chatgpt-discussions
Dall E's current inpainting needs an update. Especially when it comes to hands.
When will dalle e2 be up again?
It appears they have put pandora back into the box 😦
Do the people who buy tokens get priority over trial accounts?
Hey DALL-E-Team. Please give an update about the status of your website. I am truly thankful for what you have accomplished, but especially as a person who believes in what you do and who is ready to build his own life around your achievement, it is hard for me not to know, why the website is down 🙂
I have a question
Is creating art with AI a paid thing or free?
• You get 50 free credits your first month.
• 15 free credits will replenish every month after that, on the same day of the month.
• For example, if you signed up on August 3rd, your free credits will refill on September 3rd.
• If you joined on the 29th, 30th, or 31st of every month, your free credits will refill on the 28th of every month.
• Free credits don’t roll over to the next month, so they expire after month after they were granted – but you’ll get 15 new free credits.
• Paid credits will expire after 12 months from purchase.
How to create a new image?
You can use Modmail in the right column to send a message to the mods.
Guys can someone tell me how chatgpt is charged and what it is charged for, I have not used it for months before I registered and just started using it today
Please ask about ChatGPT in #chatgpt-discussions. This is a support channel for Dall-E.
Recommended guides are listed in #📚┃resources. A set of basic skills are covered in #1021130377026351105. User-generated content that has gotten a lot of positive reaction appears in #hall-of-fame.
Inappropriate content, such as NSFW messages or media, is not allowed on the server. Additionally, having an inappropriate profile will result in a ban from the server.
?help
you need to click the generate image button from the bot
can any one help me??
I have a problem
How to use
What's the credits
• You get 50 free credits your first month.
• 15 free credits will replenish every month after that, on the same day of the month.
• For example, if you signed up on August 3rd, your free credits will refill on September 3rd.
• If you joined on the 29th, 30th, or 31st of every month, your free credits will refill on the 28th of every month.
• Free credits don’t roll over to the next month, so they expire after month after they were granted – but you’ll get 15 new free credits.
• Paid credits will expire after 12 months from purchase.
Is there a way to take user input (at the prompt) and make the images appear in discord?
What's the credits
Anyone know if different sizes is possible?
Pressing the share button on an image we edited in Dall e 2 is it shared on the Community channel? (I mean, is it open to everyone?) Or is it only the people we send the link to see it?
I will be glad if you answer
has anyone tried to generate a picture anime style?
@vagrant zodiac Do you have any knowledge about this?
there's not a public community channel on the site, no
so only people with the link you share can see it
Okey thenks 😊
what does 30mm mean in prompts?
Hi, do you think image generation using DALLE2 is in comparison to other generators worse (less good). Although I stick to the system, I see how hard it is to create real gems. The larger competitors (I tried one on their discord server) have more possibilities to take influence on the first generated image (or the following ones), with additional technical modifiers I suppose (not sure). Will there be more features soon, or is DALLE2 in this branch on a dead track? Do you have any oppinions?
do you know, if openAI / DALLE2 team will do anything about it?
Hi, I have a question, is it legal to create for example a Mario image with DALL E 2 and use it for commercial use, what are the copyrights and can Nintendo do something?
nahh, no clue prolly not
its nintendo, they definitely will
especially if youre doing it in a commercial level
maybe its fine if its a commissioned fan art not too sure about that too though cuz nintendo is like litigiously strict about anything related to it
Ok, if I use an API where people can prompt something to DALL E can I add something to tell DALL E to not generate something that has an intellectual property such as Mario?
it can be possible for sure
Oh, it would be nice
Discussion of other AIs is permitted in the #ai-discussions channel.
Discussions of non-OpenAI models are only permitted in the #ai-discussions channel.
As a community, we value dialog about the future of AI, including alternative technologies and their associated ethical issues.
No, Dall-E has no negative prompting.
That is an example of a photographic cue. For help with photographic cues please see the guides in #📚┃resources.
For help with anime styles, please see the gallery in #anime-manga and check the style guides listed in #📚┃resources.
ok, so is there a problem if I try to earn money from pictures generated by DALL-E? Is there a problem if these pictures are Mario pictures for example?
Subject to the Content Policy and Terms, you own the images you create with DALL·E, including the right to reprint, sell, and merchandise – regardless of whether an image was generated through a free or paid credit.
While you own the images you create through Dall-E, other copyright issues may arise if you have generated a character that is subject to copyright by some other party. We are not able to give legal advice.
I see, my problem is that in my service I added a prompt where customers can write anything they want so if they write something that is not in public domain (for example Mario) I'm afraid of having legal problems because I managed to make money out of it
I suggest you speak with your attorney about that. Sounds like this may also be something to ask about in #dev-chat.
Ok, thx for your answer 👍
@unborn oak yes saw that but can you help me with img2img generation
The img2img pipeline is not a Dall-E feature. Not sure if that is what you are asking about. Dall-E does not support style transfer.
Discussion of other ai models is permitted only in #ai-discussions
Dall-e keeps getting stuck...
Yes, there are some back-end issues. For updates see #announcements.
ok ill try later thanks 🙂
how to use
Status updates on all services are available in #announcements. No need to keep repeating outages here.
Getting started guides for Dall-E are in #📚┃resources.
The same kinda goes for the legal disclaimer chatgpt has us under when we use it. They put all use on the user including copy right issue. (This is not legal advice just a observation)
DALL-E is back up now
Is there a way to save the image file at a higher resolution?
Not using Dall-E directly, no. You can use an image upscaler as a separate step. To find upscalers please use Google.
hi, i just have a question, why did dalle2 arnt free like chatgpt ? the ia is so incredible its a masterpiece but btw no money so i have to wait everytime xd
Hello! I´m new here. I´d like to create a series of 111 similar but different looking pictures. How can I do that with Open AI?
How can i get access to dalle 2 instead of dalle
Dalle is Dalle 2
This sounds like your prompt being the issue.
You are using DALL·E 2. The invite you received from OpenAI is for DALL·E 2. The first version of DALL·E, version 1, was never released to the public. Read more here.
why cant I use any of the image generation rooms? it says I am now allowed. But the last time (months ago) i was. do i need to purchase subscription? can anyone please shed a light?
Not sure I understand what you mean, can you say it another way? This server does not provide image generation services.
the dallify-1a rooms are used for image generation using /draw. i am talking about those rooms
That was done during a brief promotional period only.
anyone knows the point/meaning behind these squares?
That is the OpenAI watermark.
I was gonna be a troll and purposefully spread misinformation by saying that the squares have stuff encoded on them like your personal information. But zoomanda beat me to it
I know that dall e 2 isn't available to the public yet but is dall e 1 out there?
You are using DALL·E 2. The invite you received from OpenAI is for DALL·E 2. The first version of DALL·E, version 1, was never released to the public. Read more here.
Hi. I am looking to know how to stylize an uploaded photo without erasing. Is that possible? Example: I upload a photo of a fuzzy little kitten and I want to create 3 different styles of the photo. Style 1: water color ---- Style 2: Charcoal Painting ---- Style 3: Psychadelic digital art, but I want to preserve compositional characteristics of the original photo. Further explanation: My friend wants me to turn a photo of her cat into a sketch. Can someone point me to the best reference or kindly explain how this is done? thanks.
Dall-E does not have a style transfer feature.
We currently don't support transforming the style of an image into another style. However, you can edit parts of a generated image and recreate them in a style you define in the prompt.
I guess that explains why I couldn't find the feature XD
Discussion of other AIs is permitted only in #ai-discussions.
gotcha
Discussions of non-OpenAI models are only permitted in the #ai-discussions channel.
As a community, we value dialog about the future of AI, including alternative technologies and their associated ethical issues.
anyone have any cool outer space related prompts?
We need different aspect ratio support
Love this ai
horse
how to use
@mild thorn
no one?
Gg
you can't really do consistent styling like that with dalle
my gpt chat is not working what should i do?
You have to take care about the inputs you provide. In addition there are OpenAIs internal technical filters, which restrict content. PROBLEM IS: There is no official list, you have to try or stick to the community to find out what is allowed and what isn't. I HAVE MADE MY Experiences. // Copied from OpenAIs Terms of use:
(c) Restrictions. You may not (i) use the Services in a way that infringes, misappropriates or violates any person’s rights; (ii) reverse assemble, reverse compile, decompile, translate or otherwise attempt to discover the source code or underlying components of models, algorithms, and systems of the Services (except to the extent such restrictions are contrary to applicable law); (iii) use the Services to develop foundation models or other large scale models that compete with OpenAI; (iv) use any method to extract data from the Services, including web scraping, web harvesting, or web data extraction methods, other than as permitted through the API; (v) represent that output from the Services was human-generated when it is not; or (vii) buy, sell, or transfer API keys without our prior consent. You will comply with any rate limits and other requirements in our documentation. You may use Services only in geographies currently supported by OpenAI.
You write a program/script and use the API.
Yes, we want also more control about image generation process.
lord shiva
Oi
ok I see, so I can write the same thing in my Terms of Services?
I tried to find a software/website which can, with image recognition technology, detects the potential trademarks in a picture so I can delete these pictures but I didn't find
appreciate the honesty
Well it would be easiest to integrate OpenAIs legal information into your terms of use -> but here I would also suggest to use a lawyer (I am no lawyer). Second approach would be to implement a type of forwarding service only - and have your users register with OpenAI directly (so your users have also individual OpenAI accounts). /
This is also of interest for me -> do you know / somebody knows which additional information is embedded into the image data (either meta data, or within the image data itsself?). From a technical point of view there is a lot possible.
I see, I think I will contact a lawyer, thx for the answer 👍
How do I make art
visit the official site of dall-e 2
this is in #1021130377026351105 but when i do the same promp it looks nothing alike
help
ok thank you!
Does anyone know how I edit the words to English?
So there's an issue? Does it normally come out in English?
i am confused, is dall-e 2 only available in hosted form
or is it possible to run locally
Yes - on the servers of OpenAI.
when does the free trial end? and what is the limit of the free trial?
What other words are there that can keep the quality of a portrait-like photo but the characters aren't standing that close to the camera?
Is there a way to combine Text along with a Picture to generate an image?
DALL·E 2 is not engineered to properly generate text. While it may get a few simple words right now and then, that is the exception rather than the rule. If you need to add text to your image, consider using an external image-editing software.
You get 15 free credits each month
• You get 50 free credits your first month.
• 15 free credits will replenish every month after that, on the same day of the month.
• For example, if you signed up on August 3rd, your free credits will refill on September 3rd.
• If you joined on the 29th, 30th, or 31st of every month, your free credits will refill on the 28th of every month.
• Free credits don’t roll over to the next month, so they expire after month after they were granted – but you’ll get 15 new free credits.
• Paid credits will expire after 12 months from purchase.
okay, i thought the name implied open source
no worries
Am I allowed to send the same picture into different channels?
Please avoid spamming images to multiple channels. The gallery channels are listed by subject.
Nice picture, but in the future please post images to an appropriate gallery channel. This is a text discussion channel.
And what if the picture incorporates various subjects?
a question, guys : I am creating images with Dall-E and in the bottom right corner it produces this multi color bar - how can i get rid of that please?
Please pick the channel that you think is the best fit for the image overall.
That is the DE2 watermark. It is produced on all images generated via the web interface, but in some API cases it may not be there. You can remove it via cropping, or by using your browser's developer mode to select a raw version of the image in cache (which personally I find awkward to do).
how do I delete images stored in the history tab?
Thank you understood
You would need to contact support@oai.com for that.
@unkempt gyro good idea and I bet they’re working on it. But because you’re turning text into images it’s a whole order of magnitude more difficult than just having text
how to get dell e 2 access??
you should click on the bot's generate image button then
How exactly does one start using dall-e? open to pointers ✊🏾
cant buy credits, is anyone else having this issue?
i keep wasting credits because nothing generates
Please fix this. I entered fire and nothing came out
Hi can I set my default color scheme for digital art?
like I don't want to use current colors in generated digital arts
you need to erase the places where you want the prompt to be generated
Does anyone know if you can create custom dimensions in dall-e?
does anyone know how I can get the 90s anime style right? like this?
it always either generates something way too painting-y or something completely unrelated
I think I should ask this in #❌┃prompt-help
You have to use dall e not chat gpt
hi is there a way i can get rid of the muilti color box on the right bottom corner?
ig thats the identification of that. ig it cant be removed
its like a watermark
Open AI don’t have a problem with you removing the watermark at this time. It’s your picture. See the conditions.
I can't sign up for dalle...
can anyone help?
Can someone explain the difference between de and de2, please? And specifically the difference in accessing them? Since de2 is supposed to be so much better, I'd like to try it... but if I try to log in from the de2 start page, I get a blank white page, whereas login into de works fine...
And why did this bot message pop up just now? I wasn't trying to generate an image... btw, the link it provided is just what I described above, a blank white page..
Sigh
Is everybody seing the bot, or just me?
https://labs.openai.com is dalle 2. Dalle 1 was never released to the public
So you're saying if I was logged in somewhere and have created images, then I was using de2. In the top left it sais DALL-E, but ok. That was yesterday. Tiday I can't s3em to access qny of it anymore. Every link just leads to that white page. I can login at openai.com, but that looks completely different. It's all very confusing.... but I guess it may have to do with me accessing via my Android tablet. Will try again f4om my desktop computer when I get home...
might be due to your phone yes. try clearing cache.
Also, 'Dall-E' is the branding, 'Dall-e 2' is the model that's used on the website.
Nice picture, but in the future please post your images in the gallery channels, not this discussion channel.
Discussion of other AIs is permitted only in the #ai-discussions channel.
Discussions of non-OpenAI models are only permitted in the #ai-discussions channel.
As a community, we value dialog about the future of AI, including alternative technologies and their associated ethical issues.
Attention Off-topic chats have been cleared by a moderator.
Will Dall E be able to add things to existing images you give it?
Hi, how can I know my image credit count ?
on dall e page
when you click on your account icon you can see it in that small panel
like this
thanks, I've been blind 😅
How to generate realistic African images ..who's got more descriptive prompt for that.
The images I'm generating are not good enough
Any help?
Discussions of non-OpenAI models are only permitted in the #ai-discussions channel.
As a community, we value dialog about the future of AI, including alternative technologies and their associated ethical issues.
has there been some tuning done to DALL-E2? I am not getting stuff that is close to the work of the artists I use to describe an image, which was not how it was until yesterday.
We have not been given any specific guidance regarding possible changes to the model.
Hi all, new to DALL E 2. I have been messing about for a couple of days and am wondering, is there a way to use a picture and a prompt as input together?
How do I add image for it to edit
No, that is not an option. The exception is use of the editor functions.
The Microsoft Designer product uses DE2 on the back end.
will there be dall e 3
can you have the AI generate content from an existing image? if so how do I trigger that? for example take a photo and apply a steampunk theme to it
Discussion of other AIs is permitted only in #ai-discussions, perhaps take it up there. I don't know of a photo editing product such as you describe, other than the DE2 editor itself.
Discussions of non-OpenAI models are only permitted in the #ai-discussions channel.
As a community, we value dialog about the future of AI, including alternative technologies and their associated ethical issues.
DE2 does not support style transfer, which is what you described.
We currently don't support transforming the style of an image into another style. However, you can edit parts of a generated image and recreate them in a style you define in the prompt.
Yes, I think this was announced on February 1. This channel is for discussion of the Dall-E model, not GPT. Also, the video could be considered promotion of a TikTok channel, so it has been removed. Consider discussing the news in #ai-discussions.
You can discuss the news, but promotion of links is not permitted due to server rules. The topic pertains to ChatGPT, not Dall-E, so it is off-topic for this channel. Consider using #ai-discussions (the integration is with GPT-3.5).
In the future, please use the gallery channels to post your art. This is a discussion channel.
not sure where to post this but I was just using dalle, generated an image, went to edit it with the tool in dalle and it says that it violates the content policy. but the image was generated from dalle itself, and shouldn't violate the policy for any reason
was a weird glitch, figured I should report it
I have seen some problems with uploading images before, with false positives. Not sure what the upload filter is seeing sometimes. I would not worry about it.
what's weird is it wasn't an uploaded image, like the flow was
- put prompt in
- generate 4 images
- notice hands look off in one but everything else was good
- toss it into edit to try to fix them
- notification about content policy
the prompt I was using specifically was "gold miner from pioneer days during westward expansion, antique sepia film photo" and there was a pickaxe in it, I assume maybe that triggered something regarding weapons?
Yeah i used dall e on my phone and saved an image (it went to files) so i had to save it from my files app to my camera to send it here but i dont know if that changed the file’s formatting and it didnt work
When you say toss it in edit, you mean you selected one image and then used the edit functions, right? No download.
yeah, I selected an image and used the edit function in dalle. no edits outside of dalle whatsoever
The reference to the pickaxe would have been caught during generation. Let me test the prompt for you. One sec.
if it matters, I've done like a few hundred generations with that prompt over the last couple days and this is the first issue I've encountered. have had a bunch of other generations with pickaxes in them that worked okay
No problem with the prompt. Now I will test an edit from the set.
so it's not like it happens often, I just thought it was strange it flagged something created fully in dalle as violating the content policy. I can just generate more images so no worries on that part, but my immediate concern was potentially having my account suspended or something for content violations from something that dalle made itself haha
No problem editing the one that had the most obvious axe. I have the same problem you had now and then, impossible to know what the AI is seeing that it is not happy with.
As we're ramping up DALL-E access, safe usage of the platform is our highest priority. Our filters aims to detect generated text that could be sensitive or unsafe. We've built the filter to err on the side of caution, so, occasionally, innocent prompts will be flagged as unsafe.
Although suspensions are automatic, we manually review suspensions to determine whether or not it was justified. If it wasn’t justified, we reinstate access right away.
If you have any questions on your usage, please see our Content Policy.
No worries. If you have editing issues in the future, you can use #❌┃editor-help for more discussion. Nice art effect from the prompt, btw. Consider posting your best work in #photography since you are working with antique photo effects!
Will make sure and thanks for appreciation.
how do i create something here?
See the message.
hello is dall-e free? Like it doesn't include any sudden free trials the more I generate them?
Why can't it draw the way I want?
describe it more in detail.. and if you want that the AI draws it write any like "draw....".
you have to paste this on the dall-e page..
Do you have a account?
yes. I do
see.. m not able to generate images like I have seen in the channel of general art. it is providing me some basic images
very basic, I mean
Take a look #❌┃tips-n-tricks - there are good tips! 🙂
okay! thanks! will go through this!
No problem and good luck. 🙈
Can dall-e finally generate faces that arent disfigured?
any tips to generate cool images, i feel inferior seeing these beautiful pieces
How to use dalle
just type what you want it to generate
Is there a limit?
there shouldnt be, you can generate the images until your credits dont run out..which dont run out easily at all
Can you give me a description of making good graphics photos?
you can check out #1021130377026351105
This server fr has over 2 Million members..
Dall-E 3 is what we need
Does DALL-e save the images you sent to alter
How can I get it to make a drow elf? Whenever I try it makes pictures of brown skinned humans. They're cool looking but not what I'm after when I'm trying to get a fantasy evil elf.
the dnd guide in #📚┃resources might help?
Oooh! I'll look into it, thank you!
this is the channel for dall-e (art generation), not chat gpt
How does everybody keep getting these really nicely detailed pictures? do i have to add a key word? i keep getting really messed up faces and wobbly contures. can anyone help ?
Are there any channels for something like a caption contest?
For guides to prompt design please see the guides in #📚┃resources.
Attention Off-topic chats have been cleared by a moderator.
Why do my pictures get removed without any explanation from the moderators?
Try using ‘portrait photo’ or even ‘professional portrait photo’ in the prompt. For more details see Resources tab, there is a whole section on improving photography.
Try using ‘portrait photo’ in the prompt, also don’t mention too many other things, just concentrate on the person.
See #📚┃resources message (it’s in the Resources section)
Sorry everyone, made a bit of a hash of that! Was really only trying to reply to “wgrool”! As you can probably tell it’s a long time since I last used chat rooms… (I think it was probably when you still had dial-in modems…!)
Why is Dalle2 so bad at biblical scenes?
hey any new features in dalle since outpainting?
all i see is chatgpt in announcements
hey ! how to use dall-e-2 ?
Gm ☕️
Can DallE “photoshop” an image? Could it for example remove wrinkles from a headshot?
hi yall, do you do the art or is OpenAI creating it?
I applied for volunteer a while ago and havent got a single messege and I belive a passed it
Anyone know if dalle 2 can do text that makes sense instead of jumbled up ?
why does dall e struggle to generate images containing text without jumbling up the letters
Only if you can scribble with a fat pen, and describe exactly what you want
DALL·E 2 is not engineered to properly generate text. While it may get a few simple words right now and then, that is the exception rather than the rule. If you need to add text to your image, consider using an external image-editing software.
Dall-E has an editor function, but it is not desiged for the type of editing you would do in a digital art product. It generates new content from text prompts. To learn about the editor, see the guides listed in #📚┃resources.
Nice picture, but in the future please post your art in the gallery channels, not here in the discussion channel.
Images may be removed at any time if they appear to violate the TOS. Please read and comply with the content policies.
• Terms of Use
• DALL∙E - Content Policy
• API Terms & Policies - Service Terms, Sharing & Publication Policy, Usage Policy, Terms of Use, Privacy Policy
Posting non-DE2 art in DE2 channels is not permitted, and images must comply with the content guidelines for individual channels.
I know that but they didn't seem to violate the TOS. They were pictures of a pregnant woman and a pregnant man.
Where can I use this service?
my credits refill in 3 hours
Pls send me the link of the Dall E
The link of chat GPT
I don't have more credit what to do ?
Dall-E 3
any issues with Dall e it aways opens with error something went wrong
this channel is for discussing Dall-e. Discussions of other AI goes in #ai-discussions
Bro i out of credits ):
Does Davinci API model contain DALL-E2?
Not sure which prior images you are referring to. Your current pictures on that theme in #photography are there
Thanks @velvet karma It would be useful if moderators could let users know why pics have been taken down so they don't make the same mistake again. I appreciate your help though 👍
No problem. Don't hesitate to ask in the future. I don't know what images of yours were removed without doing an account review. Sometimes things are off-topic for the channel they are in or other reasons. When pictures are removed as part of a warn, which is more serious, you should always see a clear reason in those cases.
Attention Off-topic chats have been cleared by a moderator.
Quick question, is there a why to come in contact with a sales person from OpenAI regarding Dall e? Preferable a phone number.

