New Crazy-A** Software Takes Internet by Storm: DALL-E Mini

If you’ve been on Twitter lately, you’ve noticed that people have been sharing all sorts of crazy images. Like, crazier images than usual.  

You’ve probably seen users combine so many words and phrases to generate these insane pictures. A rabbi holding an avocado?  A group of police officers interrogating a jar of pickles? Why? How? What’s happening? The images in question can be anywhere from surprisingly realistic, or hysterically disturbing. 

If you’re wondering how people create these images, you’ve come to the right place. Let’s take a look at the newest AI software that’s sweeping the internet; DALL-E Mini.

Where Did This Even Come From?

First off, you may be confused as to how these images were even made. Where the hell did they come from? Well, it can all be tracked down to two culprits: Google Imagen software along with OpenAI. 

OpenAI is a startup that was backed by Microsoft. They are currently in the process of creating this new software called “DALL-E 2.” It is beautifully described as “a new AI system that can create realistic images and art from a description in natural language.”

In plain English, DALL-E 2’s objective is to generate images from text. 

Now, unfortunately, this isn’t available to the general public. Not yet, at least. As for now, the only way to gain official access is to join a dreaded waitlist. You also have to prove that you are some kind of professional designer, illustrator, or in a related field. 

The images you’re seeing likely are from  DALL-E 2’s early users. Often friends or relatives of the employees work for OpenAI. Or, you’re seeing images generated from another system; DALL-E Mini.

The DALL-E Mini Tool

DALL-E Mini is an open-source code available to the general public. This was created by a different, smaller team of loosely organized developers. As you can probably guess, it’s a result of the constant demand and impatience for DALL-E 2’s release. 

The process is rather simple. You can visit their website at any time, and type a word or phrase into the text bar. After that, you can press “Run” and the tool will create 9 images for you. 

 Much like its parent tool, these images are generated by pulling from multiple online sources. It identifies key parts of your text, then finds various images from across the web. This includes Flickr and Google Images, to name a few. 

Due to the website’s popularity, things can get complicated on the technical side of things. Demand overload will cause users to often get a pop-up when running a prompt. “Too much traffic, try again later.”  has become a notorious staple of the website. Though because the demand will almost always be overloaded, it’s best to not actually try again later. Instead, you want to keep clicking “Run” until the message disappears. 

Yeah, things can get annoying like that sometimes. But in the end, you get these really sweet and hilarious pics to show off to your friends. I’d say it’s worth it. DALL-E is a simple and fun time-waster for the average user. And a future super-serious tool for professional designers. 

If you’d like to try out the new technology for yourself, here it is. Be warned; you may get addicted.