Using DALL-E 3 Directly in ChatGPT for Image Creation

In the ChatGPT image upload post from yesterday, I mentioned that a direct connection to DALL-E 3 was not available, so image creation and download were not possible... However, today I noticed that DALL-E 3 has become selectable in the new option menu.

DALL-E 3 Beta visible at the very bottom.

Now, you can directly ask ChatGPT for image creation and even download it after creation. It's truly convenient. Let’s delve into more details below!

@ Quick note~ What is DALL-E 3?
It’s a text-based AI image creation engine from ChatGPT, capable of handling various requests related to images.

Trying Out DALL-E 3 in ChatGPT

Initially, I was unsure what to try, but I wanted to create a new version of the globe icon in the header of the Webisfree.com site. So, I made a request for icon creation as follows:

The result was a message saying that only large images could be created.

I immediately requested for creation, and soon after, I was able to see four amazing images created. It's truly simple and astonishing!

Image created using ChatGPT's DALL-E 3.

Personally, I think the first image is the most suitable for use. However, I would prefer an image using only black and white. So, I requested for it to be recreated.

Requesting Image Recreation in ChatGPT

It is possible to comfortably request image recreation in ChatGPT. Initially, I requested the first image that I liked the most to be recreated in black and white as follows:

Immediately, an image was created as follows. Creating the needed image on the spot is incredibly convenient.

Newly created black and white icon upon request.

Creating the desired icon easily and quickly was very convenient. Now, I only need to make good use of it.

@ A sudden thought... Is it possible to create Gif images?
I suddenly wondered if it was possible to create moving images and thankfully, a moving gif image was created this time. However... I thought, "now it can even create moving gifs", but upon actual download, it did not move... Upon asking again, I was told that creating moving gifs is challenging with DALL-E 3. It’s unfortunate, but it's not a feasible request at the moment.

Next time, it will create moving gifs too. Then, it seems like it won’t be long before it can create moving videos!

In Conclusion

Today seems like it was genuinely a delightful experience. I’ve tried text-based image creation through various other apps, engines, or through bing.com AI before, but obtaining the desired text and image all at once with ChatGPT is truly astonishing and can be expected to bring about significant productivity improvements. I will make sure to update quickly on any changes that occur in the future.