Just a few days ago, OpenAI announced the release of Dall-E 3 to all ChatGPT Plus users in early October.

The new version of Dall-E "understands significantly more nuance and detail than our previous systems, allowing you to easily translate your ideas into exceptionally accurate images." (Source: https://openai.com/dall-e-3)

Great news for all inpatient AI artists like myself - we don't have to wait until this release and don't have to pay for ChatGPT Plus to start making images with Dall-E 3!

You can use Dall-E 3 with Bing image generator for absolutely free.

Just sign in with your Microsoft account to Bing Create and start creating unique artwork! 

You will get 100 boost credits initially, making your image generation faster. After you use them all, image generation will still be possible, but slower.

Test it with a prompt, which includes written text in the desired image.

Try something like: "A grey wolf in the forest holding a sign that says: "Hug Me!"

Similar prompt: A grey cat with green eyes is holding a sign with the text: Feed Me!

Try also the "colorful illustration" version of this prompt. Which one do you like more?

This image I created for a contest on X/Twitter with a Dragon theme; it got the most likes so far:

"Beautiful orange dragon in a forest holding a sign that says: "I'm on fire!" photorealistic. Dry grass and trees are burning behind him."

You can also generate now your Logo or artist name in new ways:

"Artistic illustration of the Words: REAL STELLA SKY; colorful flowers, Art Deco"

"A fantasy illustration of a beautiful woman with short red hair with a black T-Shirt with white text: Stella Sky. Futuristic City park like Sky Gardens in Singapore."

The text is not always perfect; sometimes, you have to make several generations to get a great result. The rest was accurate, and the city park looks like Sky Gardens in Singapore, which I visited years ago.

I share free prompts and tutorials on my blog, and I tested the same prompts in Dalle-E 3 and Midjourney in --ar 1:1 format for better comparison.

It is difficult to say which images are better; it depends on subjective preferences.

Free Midjourney Prompt "Beautiful Lady in green bubbles!

"real woman with very short grass-green hair, green eyes, soap bubbles, golden lights, fantasy world, magical lights, Art Nouveau, airbrush illustration, --ar 1:1"

Psychedelic colorful illustration, Art Deco illustration, a beautiful woman, short red hair, green eyes, pale skin, feminine body, a beige lace dress, on a white-sand beach, DeepDream --ar 1:1

"A fantasy illustration of a beautiful woman with short red hair with a black T-Shirt with white text: Stella Sky. Futuristic City park like Sky Gardens in Singapore."

However, overall, I like the possibilities in Midjourney to adjust your prompts, make variations of the regions, and many other features. Also, with Dall-E 3, it is not possible to use an art style of a living artist or generate an image of a celebrity, which is possible with Midjourney.

Dall-E 3 is, for me, a winner regarding images with text. Midjourney needs to catch up in this department; I hope they will improve soon.

"When prompted with an idea, ChatGPT will automatically generate tailored, detailed prompts for DALL·E 3 that bring your idea to life. If you like a particular image, but it’s not quite right, you can ask ChatGPT to make tweaks with just a few words." (Source: https://openai.com/dall-e-3)

I will test the image generation with the help of ChatGPT Plus and let you know on my blog if it is worth the subscription.

Subscribe to my newsletter to get the news quickly and receive new prompts!