Create multiple consistent characters with dall-e 3 & Custom GPT

AI Money Maker
20 Jan 202408:01

TLDRThe video introduces a method for creating consistent characters for various creative projects using a custom GPT model. By establishing detailed parameters and using a base prompt, users can generate characters in a 3D Pixar style with unique attributes, such as a neon aura. The process involves fine-tuning the prompt, generating images, and selecting the best results to maintain character consistency. The video also discusses enhancing image quality and incorporating multiple characters without losing consistency, ultimately leading to high-quality project outputs.


  • ๐ŸŽจ The video introduces a method for generating consistent characters for various creative projects such as storybooks, animations, and comic books.
  • ๐Ÿ‘พ The presenter has achieved the best results to date using an art generator to create animations and comic book pages with consistent character styles.
  • ๐Ÿ’ก To create a custom GPT, one must subscribe to a GPTs Plus plan for $20 a month, which allows for the creation of custom GPTs and image generation using Dolly.
  • ๐Ÿ“ The process involves configuring a GPT by providing specific details about the characters, including style, appearance, and any unique attributes.
  • ๐ŸŒŸ The video suggests using up to three main characters without losing consistency, as the AI may get confused with more.
  • ๐Ÿ–Œ๏ธ The presenter emphasizes the importance of a detailed base prompt for each character, which can be refined by generating and reviewing images until the desired look is achieved.
  • ๐Ÿ”„ Once satisfied with the character design, the base prompt should be saved and used as a reference for future image generation to maintain consistency.
  • ๐Ÿ“ธ The video mentions the use of Dolly for image generation, but notes that the images are low resolution and may need to be upscaled for commercial use.
  • ๐Ÿ–ผ๏ธ For projects built within Canva, images larger than 25 megabytes may need to be resized using free tools like photo P to meet the platform's requirements.
  • ๐Ÿ’ฌ The presenter encourages viewers to ask questions in the comments and offers to create more content based on viewer interest and engagement.
  • ๐Ÿ“ˆ The video concludes by highlighting the potential of using custom GPTs not only for personal projects but also as a means to generate income from Open AI.

Q & A

  • What is the main topic of the video?

    -The main topic of the video is a method for generating multiple consistent characters for use in storybooks, animation projects, comic books, or other creative projects.

  • What are some benefits of using this method for character creation?

    -Using this method allows for the creation of highly consistent character designs across different scenarios and scenes, which can be crucial for maintaining a cohesive visual style in various projects.

  • What is the role of GPT in this character generation process?

    -GPT plays a crucial role by allowing users to build their own custom GPT, which can then be used to generate images of characters based on specific parameters and styles provided by the user.

  • What is the cost associated with creating a custom GPT for this purpose?

    -To create a custom GPT for generating images, users need to upgrade to a GPTs Plus plan for $20 a month.

  • How does one begin to create a custom GPT for character generation?

    -Users start by going to the explore tab and then creating a GPT. They then proceed to the configure page and input the specific parameters and styles for their characters.

  • What kind of information should be included in the initial character description?

    -The initial character description should include as many details as possible, such as name, age, hair color, eye color, clothing style, skin color, and any other distinctive features.

  • How can users refine their character prompts for better results?

    -Users can refine their character prompts by generating an image with the initial description, reviewing the GPT-generated prompt, and then adjusting the details to remove any unnecessary explanations and focus on specific character traits.

  • What is the recommended limit for the number of main characters in a project using this method?

    -It is recommended not to exceed three main characters, as the AI may begin to get confused with more characters.

  • How can users maintain consistency in character design across multiple scenes?

    -Users can maintain consistency by using the character's name and a description of the scene when prompting the GPT. Additionally, they should save the best and most similar images to the bot for reference.

  • What is the process for incorporating a second or third character into the scenes?

    -To incorporate additional characters, users should repeat the process of creating base prompts and generating reference images for each new character, ensuring that the AI has all the necessary information to maintain consistency.

  • What is the recommended method for upscaling low-resolution images for commercial use?

    -Users can use an upscale AI image upscaler like Photo P to increase the resolution of the images. They should select the 'general photo fast ra' setting and can also batch upscale images for efficiency.



๐ŸŽจ Introducing Custom Character Generation

The paragraph introduces a method for generating consistent characters for various creative projects such as storybooks, animations, and comic books. The speaker shares their excitement about the results achieved with an art generator and provides examples of animations and comic book pages created using this method. They also mention a base prompt provided in the video description to help viewers create their own custom GPT for character generation. The speaker encourages viewers to like the video for more exposure.


๐Ÿ–Œ๏ธ Configuring GPT for Character Consistency

This section provides a step-by-step guide on how to configure a custom GPT to generate consistent characters. The speaker explains the process of naming the bot, filling in specific information in the prompt, and adjusting parameters such as style, aspect ratio, and character details. They also discuss the importance of creating a detailed prompt for the character and refining it through multiple iterations. The speaker emphasizes the limit of three main characters to avoid confusion and suggests using reference images to maintain consistency.



๐Ÿ’กArt Generator

An art generator refers to software or a platform that uses artificial intelligence to create visual art based on input parameters or prompts. In the context of the video, the art generator is used to produce consistent characters for various creative projects such as storybooks, animations, and comic books. The narrator emphasizes how the method they share enhances the consistency of characters across different scenes and scenarios, showcasing the technology's capability to maintain style and appearance seamlessly.

๐Ÿ’กCustom GPT

Custom GPT indicates the process of creating a personalized version of the Generative Pre-trained Transformer, an AI developed by OpenAI, tailored to specific needs or projects. In the video, the creator describes how viewers can build their own custom GPT to generate images that are consistently styled. This process involves configuring the AI with detailed instructions that align with the creator's artistic vision, ensuring the output matches the desired consistency and style for their characters.


Consistency in this context refers to the uniformity and coherence of character designs across different illustrations or scenes within a project. The video highlights the importance of consistency in creative works like animations and comic books, ensuring that characters retain their distinctive styles, colors, and features in every image. This consistency is crucial for storytelling, helping the audience easily recognize characters and maintain immersion in the narrative.

๐Ÿ’กBase Prompt

A base prompt is a carefully crafted text input designed to guide the AI in generating specific types of images or art. In the video, the creator suggests starting with a base prompt to expedite the custom GPT configuration process. This base prompt contains detailed descriptions of characters, including their appearance, style, and any unique attributes, such as a neon aura, to ensure the AI produces images that meet the creator's exact requirements.

๐Ÿ’กImage Upscaling

Image upscaling is a process used to increase the resolution of digital images without compromising their quality. The video suggests using upscale AI image upscaler tools to enhance the resolution of the images generated by the art generator, making them suitable for commercial purposes or high-quality printing. This step is crucial for creators looking to use their AI-generated art in professional projects or merchandise.

๐Ÿ’กScene Generation

Scene generation refers to the creation of complex images depicting specific scenarios or settings, involving characters and environments. The video demonstrates how custom GPT can be used to generate scenes with consistent character appearances, even when introducing multiple elements or characters. This capability allows creators to produce a series of coherent images for storytelling, enhancing the narrative flow of books, animations, or comics.

๐Ÿ’กCharacter Description

Character description involves providing detailed information about a character's physical appearance, clothing, and any distinctive traits. In the video, the narrator outlines how to craft a detailed character description for the AI, including aspects like age, hair color, clothing style, and unique features like a neon aura. This comprehensive description serves as the foundation for generating consistent and accurate character images using the AI.


DALL-E is an AI program by OpenAI capable of generating images from textual descriptions. In the script, DALL-E is mentioned as the tool used for image generation, requiring a subscription to access enhanced features. The video showcases the use of DALL-E to create visually consistent characters for various creative projects, illustrating its effectiveness in bringing imaginative concepts to life with high fidelity to the creator's vision.

๐Ÿ’กReference Image

A reference image is a picture used as a visual guide to aid in the creation of new artwork. In the context of the video, reference images are uploaded to the custom GPT to serve as examples or standards for generating new images. These reference images ensure that the AI maintains consistency in character appearances by comparing new outputs with the established visual benchmarks.

๐Ÿ’กGPT-Plus Plan

The GPT-Plus Plan refers to a subscription-based model offered by OpenAI for accessing advanced features of their GPT (Generative Pre-trained Transformer) technology, including the ability to create custom GPTs and generate images using DALL-E. The video mentions that upgrading to this plan is necessary to utilize the method described for generating consistent characters, highlighting the financial aspect of accessing these advanced AI capabilities.


The speaker introduces a method for generating consistent characters for various creative projects such as storybooks, animations, and comic books.

The speaker shares their excitement about the results achieved with an art generator, claiming it to be the best they have encountered.

The process involves building a custom GPT to achieve these results, with a base prompt provided in the video description for viewers to adapt.

An upgrade to a GPTs Plus plan is required to create custom GPTs and generate images using Dolly.

The speaker provides a step-by-step guide on configuring the GPT, including naming the bot and filling in specific information.

The importance of establishing a unique style for characters, such as a 3D Pixar style with a neon aura, is emphasized for visual consistency.

The speaker explains how to refine the character prompt by adding detailed descriptions and then condensing it for effective use with the GPT.

The process of generating an image, reviewing it, and then fine-tuning the prompt for better results is outlined.

The speaker advises on not exceeding three main characters to avoid confusion for the AI.

Once satisfied with the characters, the speaker instructs on saving the base prompts and reference images for future use.

The speaker demonstrates how to test the bot by generating scenes using character names and scene descriptions, resulting in consistent character portrayals.

The speaker addresses the challenge of maintaining character consistency when introducing additional elements or characters in the scenes.

The effectiveness of the custom GPT in generating consistent characters across different scenarios and scenes is showcased.

The speaker provides tips on upscaling low-resolution images for commercial use and integrating them into projects built within Canva.

The speaker invites viewers to ask questions about the process and offers to create a dedicated video if there is enough interest.

The speaker expresses confidence in the method's ability to help viewers generate consistent characters for their projects.