These 45 awesome images were made by Stable Diffusion artificial intelligence
(Pocket-lint) – There are various tools now for creating your own images using artificial intelligence. These are powerful tools for generating surprisingly good images with just a few words or requests. We’ve seen some excellent images created by Google’s Imagen tool and with Dalle-2 as well.
Now it’s the turn of Stable Diffusion, a powerful tool which is able to craft some fantastic images that could easily have been created by artists.
stablediffusionweb
A delicious cheesecake
We’ve no doubt that these artificial intelligence tools have been shown a lot of photos of food. So Stable Diffusion should have no trouble creating some tasty treats.
POCKET-LINT VIDEO OF THE DAY
Here we asked the tool to create an image of a delicious chocolate and raspberry cheesecake. The results are certainly tasty.
Best iPhone apps 2022: The ultimate guide By Maggie Tillman . 30 April 2022
These are the absolute best iPhone apps available right now, from productivity apps to apps for traveling, reading, listening to music, and more.
stablediffusionweb
A photo of medieval knight crying in the rain
The medieval ages were certainly a hard time so it’s no wonder these knights are crying.
We like how flexible these AI tools are. Whether creating realistic images or paintings.
stablediffusionweb
Disneyland painted by Van Gogh
One of the fun things you can do with these tools is to mix two different styles or to create a new painting in the style of someone long since passed. Like here we’ve asked Stable Diffusion to paint a photo of Disneyland in the style of Vincent Van Gogh.
You can clearly see the stylings of Starry Night in the image.
stablediffusionweb
Apple’s Augmented Reality glasses
Although these images aren’t that great (what’s happening to that man’s hand?) we do enjoy the weird and wonderful visions of Apple’s AR glasses that the AI has conjured up. Certainly different to anything we’ve seen so far, but also unlikely to be a final Apple product design.
stablediffusionweb
A futuristic city in the middle of the desert
You might have seen the plans for the futuristic city that Saudi Arabia is working on. But this is what AI thinks a future city in the desert might look like.
stablediffusionweb
It’s raining cats and dogs
We were hoping for cats and dogs to be pouring down from the sky with this one, but the view of wet dogs in the rain is equally good.
We’re not sure what’s going on in the second image though, maybe the puddle is too deep for the puppy, but what’s that at the back?
stablediffusionweb
Alien robots invading Time’s Square
If aliens invaded what would it look like? No one really knows and AI certainly doesn’t either. So when you ask for images of an alien robot invasion you certainly get some weird creations.
stablediffusionweb
Brand new Banksy Artworks
Banksy’s artworks have a certain recognisable style to them so even though these aren’t real you might have to glance twice to realise that.
Though the message of the art certainly isn’t as clear.
stablediffusionweb
A high resolution photo of a british bulldog wearing a top hat and smoking a cigar
Like many people, artificial intelligence hasn’t really seen dogs smoking cigars or wearing fancy clothes. So it struggles to imagine what that might look like. Hence why we’re seeing a dog seemingly smoking out of its ear.
stablediffusionweb
A modern electric version of a retro classic car
We’ve seen plenty of weird and wonderful concept car creations made by human designers.
But with AI you can make some suggestions and see what happens. The results might well inspire the creation of real vehicles in future. Who knows what’s possible?
stablediffusionweb
A modern photo of the Girl with the pearl earring
You’ve probably seen the Johannes Vermeer painting Girl with a Pearl Earring, but what if the woman in that painting lived in modern times and was posing for Instagram?
stablediffusionweb
A deep space photo by James Webb Telescope
We’re often amazed by the incredible images captured by the James Webb Space Telescope.
So we wanted to see what the Stable Diffusion could create. These could easily be real photos of deep space.
stablediffusionweb
Dogs happily swimming in a pool
These dogs are thoroughly enjoying themselves. Though we’re not sure the AI understands what a happy dog looks like as those smiles are odd.
stablediffusionweb
Robots in an art gallery looking at paintings of cats
One thing that’s clear is you need to be precise with your requests to the AI.
We asked for robots looking at paintings of cats and we got cat robots.
stablediffusionweb
Cybernetic humans in a pop art style
There are a lot of different styles that the Stable Diffusion AI can create. If you’re after an image with a familiar vibe then just name it and admire the intriguing results.
stablediffusionweb
A sea monster destroying a fishing boat during a storm.
We think the AI might have had a struggle with working out what a sea monster looks like but it’s done a magnificent job on creating an atmospheric view of an angry ocean.
stablediffusionweb
A strong cup of tea and fish and chips
We tried to create a thoroughly British view but frankly, that tea does not look like anything we’d be willing to drink. The fish and chips are convincing though.
stablediffusionweb
Bob’s Burgers Special Episode
With this image, we tried to see if the AI was a fan of our favourite cartoon as well.
The results are interesting though we’re not sure the AI will replace the cartoonists on the show.
stablediffusionweb
A high resolution render of a male lion with fine detail on its mane
If you don’t look too closely, the detail on these AI-created lions is really impressive. The manes might be a bit wavy, but the creatures are magnificent.
stablediffusionweb
Photograph of Loch Ness Monster taken at sunset
We’ve never seen a decent photograph of the Loch Ness monster, so these AI-created ones are actually fairly convincing.
stablediffusionweb
A vibrant pastel drawing of the Loch Ness monster
For a very different view of the same creature, we asked Stable Diffusion to create a pastel drawing of the Loch Ness monster instead. Now the monster looks like it would be right at home in a kid’s book.
stablediffusionweb
A painting of a futuristic yellow submarine plane flying through the sky
This one is based on a prompt from a site called Lexica which gives useful terms you can put into Stable Diffusion for interesting results.
The prompt here was “a painting of a futuristic yellow submarine plane flying through the sky, red wings, concept art by Ian McQue, cgsociety, highly detailed, artstation, concept art, sci-fi” which as you can see is detailed and very specific.
stablediffusionweb
A juicy burger with cheese, tomatoes, lettuce and bacon
Here’s another image that shows how important it is to be specific and detailed with your requests. One of the burgers looks ok, but the other comes with a cup of lettuce leaves. Weird.
stablediffusionweb
Highly detailed digital drawing of teenage mutant ninja turtle Donatello
With the help of artificial intelligence, it’s possible to create a wide variety of different images of your favourite characters.
stablediffusionweb
Closeup of Geralt of Rivia
Here’s another of our favourite fictional characters recreated by AI.
This time it’s Geralt of Rivia from The Witcher series (of books, games and TV shows). The results you’re seeing are based on a specific prompt to ensure the right details: “closeup of Geralt of Rivia, D&D, fantasy, intricate, elegant, highly detailed, digital painting, artstation, concept art, matte, sharp focus, illustration, hearthstone, art by Artgerm and Greg Rutkowski and Alphonse Mucha”
A Van Gogh painting of an orangutan
Another image inspired by the artwork and style of Van Gogh but with a couple of really awesome looking orangutans.
Front shot of a cyberpunk gazmask robot character,
Weirdly the images of robots created by Stable Diffusion seem to be more impressive than those of people.
The detail is fantastic with these sorts of robot illustrations.
Jesus in Cyberpunk 2077
We’re big fans of the aesthetic of Cyberpunk 2077 and it’s pretty awesome that you can use the AI to also make Unreal Engine renders. Thereby creating things like images of Jesus in the game.
Highly detailed portrait of Danny Devito
Several of these artificial intelligence art tools have rules around creating images of celebrities. The fear being they could be used to create fakes that people wouldn’t want in the public domain.
This image shows a detailed cartoon portrait of actor Danny Devito. Though the likeness isn’t perfect, it’s still great.
Iron Man with a heart reactor
It looks like you can redesign your favourite superheroes using AI too. Now Iron Man looks somehow a cuter hero than normal and maybe a bit less threatening.
A portrait of Salma Hayek as a barbarian
Another celebrity gets the AI treatment.
This time Salma Hayek has become a warrior. We feel like she could pull this off quite easily too and look magnificent doing so. The first of these images really looks like her.
A red fox as a female bounty hunter in the Wild West
We can see AI being used as inspiration for the look of a character in a movie.
With some suggestions, it can craft some images that could then be used to storyboard or plan things out before production starts.
Synthwave demonic alien face with neon horns
Here’s another example of just how many different styles the Stable Diffusion AI can create and the unique artwork that comes about as a result.
stablediffusionweb/lexica
A portrait of a chubby Mark Zuckerberg hairless
Mark Zuckerberg is well known for looking a tad emotionless and often more robot-like than human. So these playful portraits of him are an interesting alternative view of the man.
The legendary island sized lion lizard
We can see the artwork created by the AI working really well for illustrations for works of fiction. This huge lion lizard could easily be part fantasy universe for example.
And the artwork is convincing enough to be painted by a real artist.
Portrait of Henry Cavill as Chris Redfield from Resident Evil
Henry Cavill is a bit of a legend in geeky circles and here he’s been transported into the world of Resident Evil, filling the boots of Chris Redfield. Looks like he’d be a good fit for the character too.
A highly detailed photograph of a Ferrari
It seems that if you ask the AI to create a photo of a real-world thing it’s pretty easy for it to craft a believable one. Probably because of all the source materials out there.
These Ferraris certainly look convincing at a glance.
A photo of Albert Einstein eating samosas
You’ve probably seen a fair few photos of Einstein. Even some fake ones. But have you ever seen him eating samosas?
Now you can via the power of AI.
Neon operator Margot Robbie
Margot Robbie has pulled off a few awesome roles in her career so far. Now we get to see what she might look like in a cyberpunk-style universe. We’d certainly like to see more.
An old man in sorrow staring at the underworld
Here’s another impressive artwork with an ultra-realistic finish and impressive styling that really grabs the eye.
Cyberpunk dogs
People seem to enjoy making visions of a cyberpunk future.
This time with dogs. Surprisingly cute and weirdly wonderful cyber dogs.
Baby Groot and Baby Yoda
This one is interesting because of how detailed it is. But it’s also not showing the full request as the original prompt was to see Baby Yoda and Baby Groot together.
You can see hints of Yoda in the smaller Groot though.
A portrait of a futuristic British longhair cat wearing soldier clothes in war scene
Imagine a world where the dominant force is an advanced race of humanoid felines. That’s what we’re getting from this image of cat soldiers. The war for catnip is serious business.
Ezra miller as the Flash
Ezra Miller is playing The Flash in an upcoming movie, so this is a nice artistic vision of what that might look like.
Sonic the Hedgehog
Another of our favourite characters gets the AI treatment.
This we see Sonic digitally painted by AI and it’s fantastic.
Writing by Adrian Willings.
