We are going to explain what it is and how you can use DALL-E Mini, the fun artificial intelligence that tries to create drawings containing a description that you indicate. I warn you that if the dream of reason produces monsters, the consciousness of artificial intelligence can produce even worse ones.
Let’s start this article by explaining exactly what DALL-E Mini is and explaining as simply as possible how it works. And then, we will go on to tell you how you can use this AI to create images and unleash your imagination since you can do it for free.
What is DALL-E Mini
There is a trend to create artificial intelligence systems capable of generating images from a description that we write. Google has a project called DALLE-2, closed to users but capable of generating very realistic images from our ideas. And then there is DALL-E Mini, which is an artificial intelligence that offers much more bizarre results.
It is an OpenAI generator whose code is open and can be found on GitHub. It has been created by programmer Boris Dayma, who developed the application as part of a Google and Hugging Face competition in July 2021. The idea behind this AI is to create an algorithm to generate images similar to Google’s DALLE, but with a smaller infrastructure and lower hardware resources.
Broadly speaking, the way the AI works is as follows: you write something you want to see, and the first step is to analyze the words in the request you have written. Then, the AI searches for those words on the Internet and looks for the images associated with them, as well as the distinctive features that make that image the definition of the word.
Then, in the last steps, the AI creates a composition with fragments of those images it has searched for and tries to make the result as realistic as possible. The images are collected from the network without any kind of filter, so the creator already warns that the compositions could contain stereotypes that go against minority groups or misplaced images.
All artificial intelligence goes through a training period, in which they become better as they are used. And that is the reason why it generates such bizarre images, because it is an algorithm that is still being trained, and will gradually improve the results.
In any case, the bizarre images it generates have made many people use it as a game, and this is ultimately positive for the developer because thousands of people using AI around the world are helping to train it.
How to use DALL-E Mini
To use DALL-E Mini, just go to its demo page open to everyone, whose URL is Huggingface.co/spaces/dalle-mini/dalle-mini. On this site, you have to look at the part where a text box appears and below it a blank space.
The only thing you have to do here is to go to the text box and write in English the description of what you want to see. Here, the more elements or words you enter, the easier it is that some of them will be left out of the final composition, but the best thing to do is to try it out. Once you have written what you want, click on the Run button to let the AI start working. It may take a few minutes to give you the result.
And after a few minutes, the web page will show you a result of a vignette with nine images that it has generated from your description. Some of these images may be closer than others to what you wanted the algorithm to create.
This post may contain affiliate links, which means that I may receive a commission if you make a purchase using these links. As an Amazon Associate, I earn from qualifying purchases.