Creating with DALL-E 3! A collection of attractive image generation tips and inspirational prompts


The DALL-E 3 image generation model in ChatGPT is more than just an image generation tool. Especially for freelancers and business owners, this tool has the potential to revolutionize various projects. In this article, we would like to introduce the co and recommended prompts for using this DALL-E 3 to create images the way you want them.

ブログ「DALL-E 3で創る!魅力的な画像生成のコツとインスピレーションを刺激するプロンプト集」のアイキャッチ画像です。


1. What DALL-E 3 is?

 DALL-E 3 is a model that combines the text generation capabilities of ChatGPT with the image generation capabilities of DALL-E. DALL-E 3 generates relevant images for prompts entered in natural language. For example, typing in “cat with red apple” will produce this image.

DALL-E 3で「赤いリンゴを持った猫」のプロンプトで生成した画像です。

 DALL-E 3 can generate images in a variety of genres and styles. For example, typing “Picasso-style landscape painting” will produce an image like this.

DALL-E 3で「ピカソ風の風景画」のプロンプトで生成した画像です。

 DALL-E 3 can also handle complex prompts that combine multiple elements. For example, if you type “A frog in a wizard’s hat gets on a plane to New York City”, you will get this image.

DALL-E 3で「魔法使いの帽子をかぶったカエルが飛行機に乗ってニューヨークに行く」のプロンプトで生成した画像です。

2. Image Generation Tips

 So what are some tips for using DALL-E 3 to generate images? Here are three key points

2-1. be specific in your prompt

2-2. keep prompts short

2-3. be creative in your prompts

2-1. be specific in your prompt

 DALL-E 3 generates images based on the information contained in the prompt. Therefore, if the prompt is ambiguous, the image will also be ambiguous. For example, if you type “animal,” you do not know what kind of animal will appear, and if you type “flower,” you do not know what kind of flower will appear. So, use specific nouns and adjectives in your prompts to clarify what you want the image to be. For example, typing “white rabbit” will bring up a white rabbit, and typing “blue rose” will bring up a blue rose.

2-2. keep prompts short

 A prompt that is too long will not convey the message well, and a prompt that is too short will lack information. So, be concise in your prompts, using only the minimum necessary words. For example, typing “ocean” will not provide enough information, but typing “a rock lapped by waves on the beach at sunset” will generate an image that matches the image to some extent.

2-3. be creative in your prompts

 DALL-E 3 can also accommodate imaginative prompts. Writing a prompt with a combination that would be impossible in reality, or an original character or story, will generate an interesting image. For example, typing in “a princess living in a castle made of chocolate” will produce this image.

DALL-E 3で「チョコレートでできた城に住むプリンセス」のプロンプトで生成した画像です。

 DALL-E 3 is a tool to turn your ideas into images. Feel free to write prompts and have fun generating images.

3. Recommended prompts

 Finally, here are some recommended prompts. We encourage you to try these prompts to see what kind of images they produce.

  • – Knight fighting a dragon
  • – Dog playing soccer
  • – Stars in the shape of pasta
  • – Children in Halloween costumes
  • – Santa Claus’ factory
  • – Sushi restaurant sign
  • – Mouse hidden inside a cheeseburger
  • – Penguins having a snowball fight with a snowman
  • – Dolls as marionettes
  • – Upside-down town

4. Summary

 These are the tips and recommended prompts for image generation using ChatGPT’s DALL-E 3. ChatGPT’s DALL-E 3 differs from previous image generation AIs in that it can now generate images more flexibly, but the basics are the same. Try to use it with an awareness of the right prompt, not too long and not too short. DALL-E 3 is still under development, but improvements are being made every day to ensure that it will continue to generate higher quality and more diverse images in the future. See you in our next blog!

Back to archive