Dall-E mini: how the AI ​​image generator makes your meme dreams come true

Dall-E mini: how the AI ​​image generator makes your meme dreams come true

Dall-E mini: how the AI ​​image generator makes your meme dreams come true

Artificial intelligence has changed the world by solving highly complex math questions, making tedious tasks easy, and most recently allowing the internet to create images of Yoda and Godzilla playing poker together.

In fact, thanks to the free-to-use platform Dall-E mini, the internet is filled with a series of bizarre images of strangely twisted celebrities, cartoon characters in composite positions, and animals dressed as aristocrats.

What is Dall-E mini and how does it work?

Dall-E mini is a text-to-image AI software that creates images from your written directions.

Dall-E mini takes its name (and much of its code) from OpenAI’s software Dall-E 2. The team behind this mini version attempts to replicate OpenAI’s results through an open-source model.

The model was trained by looking at millions of images on the web associated with their captions. By doing this, the program builds a database of what an image looks like with its formulated prompt.

This software, while impressive, is by no means flawless and can often struggle to produce a perfect image, seeming to struggle with faces especially.

© Dall-E Mini

No tip, only mandatory service charge © Dall-E mini

How to use Dall-E mini

To try out the service for yourself, all you need to do is head over to the Dall-E mini website. Here you will be asked to enter and run a prompt. This may take a while to process, so expect a wait of up to 2 minutes for your image to appear.

You’ll then get nine images that match the prompt, all with slightly different variations. You will often find that the image is blurry or that Dall-E has misunderstood your prompt. If this happens, modify the command a bit and try again.

You can also request different types of images with your command by adding phrases like ‘like a painting’, ‘in a fisheye lens’, ‘like cctv camera images’ – get creative!

Keep in mind that users often get the problem of ‘too much traffic’. If this happens, try running it again – it usually works after a few tries.

© Dall-E Mini

Transforming into a hipster guitarist © Dall-E mini

Dall-E vs Dall-E mini

As we mentioned, Dall-E mini follows in the footsteps of another more developed piece of software known as Dall-E 2, but while the two platforms use the same software, their approach and end results are very different.

Firstly, Dall-E 2 is a private service that offers access on a long waiting list basis. Dall-E mini, on the other hand, is free to use for everyone. This does mean that Dall-E mini is often not available with too much traffic and slower loading times.

The mini equivalent also struggles to match the same quality as the original. Prompts can often be blurry or of low quality. Dall-E 2, on the other hand, is largely accurate and of high quality.

The other standout feature you’ll see in Dall-E mini is a better understanding of pop culture. Characters like Yoda and Thanos or celebrities like Guy Fieri or JK Rowling will appear… often in somewhat vague but recognizable forms.

A wild west of images

One of OpenAI’s big claims with the original Dall-E system was about security. It implemented a security policy for all of its images, filtering out images that are violent, sexual, or that the team deems potentially inappropriate.

A second filter is also used to try to remove images that could be seen as political, propaganda, or offensive in a more subtle way.

Finally, right now the Dall-E 2 team is checking every image to take out all the inappropriate that sneaks in.

On the other hand, the new Dall-E mini has no limitations. While the team has issued a disclaimer about prejudice and imagery. may be displayed, no search is prohibited.

This means that images have surfaced that would not meet the security restrictions of the original model, causing a bit of a wild west of images on the internet.

© Dall-E Mini

A Ted Talk on Lasagna and Trash Nachos with Guy Fieri © Dall-E mini

The future of Dall-E mini

Like the original model it was inspired by, Dall-E mini is constantly being improved and built. The team has added a screenshot button and is working on Dall-E Mega – an improved version of the model that has major improvements in mind.

However, Dall-E mini has great competition. OpenAI is slowly opening up their software to the world, allowing more people to use Dall-E 2, and Google has now launched its own competitor Imagen – a model with a similar rigor to OpenAI’s original invention.

Read more: