Although researchers around the world have long work on artificial intelligence and content generation, whether it be text or an image, it is curious what exactly DALL-E, or rather DALL-E mini, was responsible for make known to the general public how artificial intelligence allows you to generate artistic images from nothing. Or just typing a few words.
If a few years ago we learned that it was possible to generate human faces of photographic quality, which in reality did not exist, now there are several models of artificial intelligence. able to create art with a few simple commands. An example that went viral some time ago is DALL-E, whose API is available but cannot be processed without the necessary knowledge.
But if you’ve come this far, it’s because you want to play with artificial intelligence and write random words and get computer generated images. Well, you’re in luck. The bad news is that DALL-Emini is moving. The good news is that there is already an equally viral alternative, chalk. And there are other equally interesting projects for creating images with words.
DALL-E mini says goodbye
DALL-E became known, among other things, thanks to DALL-Emini, an app or online tool that lets you create images by typing a few words. After a few seconds, two minutes at the most, several drawings or hyper-realistic images appear based on the words you entered. This is how artificial intelligence DALL-E works.
When I wrote these lines DALL-Emini It is still online, but will close its doors sooner or later. The reason is that those responsible for DALL-E asked the creators of DALL-E mini to change the name to avoid confusion. Therefore, from now on, DALL-E mini is known as chalk.
Craiyon replaces DALL-E mini
Our first recommendation for creating images from words: chalk. Basically, it’s an app that went viral, but with a different name and its own domain. easier to find online. It is otherwise identical to the original, run by the same creators (Boris Daima and Pedro Cuenca) and will remain free for now.
Use chalk you just need to write one, two or more words in the appropriate field. By clicking the button on the right or pressing the Enter key, the technology behind this application, DALL-E API, start making art from the words you entered. The process is fast and you will see results immediately.
General, nine images that you can enlarge if you press and save when you right-click (on the desktop) or hold the image with one finger (on smartphones and tablets). there is even a button Screenshot create an image containing nine creations, and thereby distribute them on social networks or keep them as a keepsake.
Latent diffusion
But besides DALL-E, there are other interesting projects that want to generate images from words or other materials. inputs using artificial intelligence. One such project is LAION, also available to anyone who can create tools based on their AI. That’s how it’s born Latent diffusionwhat might remind us of DALL-Emini.
Its action is similar. It’s free, you can use it as many times as you want and you do not need to register or provide personal details. Moreover, it is hosted in the same developer repository. To create images with Latent diffusion you will need to enter one or more words in the appropriate field and then click on the button Send. But you can customize the result by changing the intermediate parameters. first steps or Steps. The more steps, the better the image quality will be. In turn, this will take longer to create. Secondly, the size of the image, maximum 256 x 256 pixels.
You can also choose number of images which you want to create, from one to four, and a difference scale between images, from 1 to 15. By default, this means 5. Playing with all these parameters, you will get curious, amusing or disturbing images created by artificial intelligence.
NeuroGen
The third online application for creating images using artificial intelligence – NeuroGen. It is not based on DALL-E, but it also promises interesting images. first 100 images created for free. From there, every 1000 images you want to create costs $10. And like it or not, you’ll have to sign up, although you can use your Google credentials to do so.
After registering, use NeuroGen too easy. You enter words for inspiration, the number of images you want to create, and the AI does the rest. In this case, this may take a few minutes. Of course, when you’re done, you’ll receive an email notification.
The resulting images are private, but you can make them public, delete them or keep them in PNG format. As an encouragement, images must be at least 256 x 256. If you open the dropdown, you can change it to “Medium” or “Large” and thus get higher resolution images: 512 x 512 or 1024 x 1024.
Source: Hiper Textual
