DALL-E 3 is live through Bing Image Creator

53273897018_b77d73cfe8_b.jpg

53273897028_1eb643f220_b.jpg

53273897058_de8bd28837_b.jpg
 
I use artist styles a lot. Nearly every prompt. I have always gotten decent art that is usually similar in some way to the artists I name.
Today, running through a whole supply of boosts, every image Bing gave me was the same low-talent heavily lined Mike Allred style, no matter which artist I named.
Has anyone experienced anything similar today?
 
You can get bloody images sometimes if you add the word wounded before a character. It doesn't seem to get blocked most of the time.

These were made with the words detailed gouache art style in the prompt:
m09Ic4j.jpg


PGHTB5P.jpg


X50Htmq.jpg


aWUqbyW.jpg
 
Last edited:
I use artist styles a lot. Nearly every prompt. I have always gotten decent art that is usually similar in some way to the artists I name.
Today, running through a whole supply of boosts, every image Bing gave me was the same low-talent heavily lined Mike Allred style, no matter which artist I named.
Has anyone experienced anything similar today?
This policy for Dall-e 3 might've played a part:
"DALL·E 3 is designed to decline requests that ask for an image in the style of a living artist. Creators can now also opt their images out from training of our future image generation models."
 
This policy for Dall-e 3 might've played a part:
"DALL·E 3 is designed to decline requests that ask for an image in the style of a living artist. Creators can now also opt their images out from training of our future image generation models."
I am really surprised anything from The Big-N is so easy to do. Normally they chomp at the bit to protect their IPs.

Perhaps it has something to do with the difficulty of blocking that which is such a big par of the world's culture.
 
Last edited:
Tried prompts including Barret from FF7, but bing-algorithm seems to censor something, somewhere.

Capitulating moment right here..
 
I stole the idea from some artists on Youtube, but I asked the AI to draw characters based on their book description/info :

"turnaround sheet of a 16th century man with long curly black hair that fall to his shoulders, piercing sparkling blue eyes, long nose that is slightly curved at the end, high forehead that is covered by his hair, pointed chin, is tall and lean with broad shoulders and muscular arms."
d'Artagnan
4LqIMBg.jpg


"watercolor painting of tall, thin woman between 65 and 70 years of age with white snowy hair, pale blue eyes, and a pinkish wrinkled face. She is often seen carrying knitting needles and yarn, and two of her hobbies are bird watching and gardening"
Miss Marple
rj1eZFx.jpg


"full body ink drawing of a 19th century man in his 30s more than six feet tall, very lean, with piercing eyes and a thin hawk-like nose. His voice was high and occasionally strident. his eyes are gray and he has a narrow face and black hair"
Sherlock Holmes
flZxb5i.jpg
 
OIG.Ue1XOvQfTKTScAjUBYcu

[/spoiler]
It wasn't really sure what to do. Her ear is kinda messed up, a leg merges into the bar stool leg, and there's another set of legs under the stool from as if she's standing. In behind there looks to be another set of deformed stool legs.
 
Last batch for today.

gaXu5Fj.jpg

oKLtFVv.jpg

pATcL4B.jpg

ZjhfS4T.jpg

I was inspired by the style of your images and decided to play around with it.

This image reminded me of a manga and anime "Record of a Yokohama Shopping Trip" I read and watched.

OIG.qiZ57.XT69foQBuJHBGQ


In that world an ecological disaster happened which caused the worlds water level to rise and the decay of humanity. Within it were scenes of neighborhoods submerged in water. This gave me the idea to create such scenes. Image Creator seems to have difficulty separating the flooded from the dry land. The road always seems to be wet even if not submerged. Might simply take some practice. At least it does a good job showing some of Image Creator's limitations.

OIG.r5ZI02fIXIfzDMr.QqMF


OIG.EuUtyoAqumEylbITNVIt


OIG..ULSAhE0qecYmJ7SKpSz


OIG.QObURznpxAjtJo.QGEAi


OIG.OTCYVJAO6eVQkegaRJXa


OIG.ICkpjWFusr3Lcm1pjGKW


OIG.e1tlQJiKxK3PHjQPWoVn


OIG.Z.4C3KpH7twvcTYl7jiz


OIG.I0v.gyJaPy4UQbEc8Zn_

Here I tried experimenting by removing the anime filter.

OIG.pYujxeXjOBYt8Tde9RsR


OIG.HlgG4KFrtQw1bTEeOw3e


OIG.nv7aDKCv6T8UXxMdvuKG


OIG.rNlG3MsJ.3wtD2Uq2kjd


OIG.l88Lq6eOGYOU4uQHW8b3


OIG.HVUBKFyb9twMLfJa_sUw


OIG.r2XEuruyRwO.YnpkzVYl
 
OZiT3yo.jpg

sOpyIdo.jpg


I still can't do a decent Wonder Woman.
Superman in a speedo with a giant bulge? No problem.
Anything remotely similar to Wonder Woman? "Unsafe Image!"
 
Top Bottom