If you're into artificial intelligence or artificial image generation techniques, I'm sure you've heard of OpenAI's text-to-image generation tool DALL-E. Although the company recently unveiled its next-generation DALL-E 2 model, the tool is limited to vetted partners and researchers. So now a developer has created a similar AI-powered text-to-image generation tool for regular users and it works really well. Check out the details here!

A developer named Boris Dayma and his team have built a DALL-E-like text-to-image generating AI tool, called DALL-E Mini, which can turn any text sentences into artistic and quirky images that may or may not make any sense. You can type whatever you want in a text box and the tool will crop relevant images to match it, much like how the original GPT-3 based DALL-E model works.

I tried the tool myself to see how it works and I was surprised by its effectiveness. Although it is worth noting that the generated images are not accurate and contain intentional blemishes and distortions to indicate that they are AI generated images and not real! You can check out a couple of examples I tried on the DALL-E Mini in the image below.

Now, after Dayma created DALL-E Mini, the tool became quite popular among users who want to create funny pictures or memes to share on social platforms. The tool is also used to create artistic placeholder icons for game objects or combine fantasy characters to create a new one. You can find many DALL-E Mini images shared by users on platforms like Twitter and Reddit.