Using Stable Diffusion (In 5 Minutes!!)

Royal Skies
29 Sept 202204:23

TLDRThe video script introduces viewers to the official stable diffusion site for AI-generated images, emphasizing its ease of use and accessibility for the average user. It highlights the site's features, such as the image dimension controller, CFG setting for prompt adherence, and steps for image refinement. The video also touches on the image editor's capabilities and minor glitches, offering tips on how to use the platform effectively for creating and editing images, including image mutation using image opacity.


  • 💻 The video series utilizes the official Stable Diffusion site for AI image generation, emphasizing support for the developers and accessibility for the average user.
  • 💰 Purchasing credits on the site financially supports the developers, enabling further improvements to the AI.
  • 💻 The site is praised for its streamlined design and user-friendly interface, featuring a dark theme.
  • 💎 A unique 'weapon height' slider allows users to adjust the dimensions of the generated image, accommodating various uses like wallpapers or mobile backgrounds.
  • 🔮 CFG (configuration) settings determine how closely the AI follows the user's prompt, with a default value of 7 offering a balance between accuracy and creativity.
  • 🔧 'Steps' control the diffusion process's duration, affecting image quality and generation time.
  • 📸 Users can choose the number of images generated per request, with options ranging from one to nine.
  • 🔬 The function of the 'sampler' setting is unclear, but several types are available for experimentation.
  • 📚 A glitch specific to Firefox affects the image editor, recommending Google Chrome for full functionality.
  • 🖋 Image editing features include resizing, panning, erasing, and restoring parts of images, with various controls for brush size, sharpness, and strength.
  • 🔍 Image opacity and mutation tools allow users to modify images further, with transparency levels influencing the intensity of mutations.

Q & A

  • What is the main reason the speaker chooses to use the official stable diffusion site for their series?

    -The speaker chooses to use the official stable diffusion site because they love what the AI stands for and want to support the developers. They mention that purchasing credits on the site directly funds the developers, allowing them to improve the product for everyone.

  • Why does the speaker emphasize the importance of accessibility for the average user in their series?

    -The speaker emphasizes accessibility because they recognize that the majority of people do not have custom-built PCs, knowledge of GitHub or command prompts, nor the time or resources to train AI locally. By sticking to the official website, the series can be more inclusive and user-friendly.

  • What is the purpose of the 'weapon height slider, controller' mentioned in the script?

    -The 'weapon height slider, controller' is a tool that allows users to change the dimensions of the image generated by the AI. This helps users tailor their images to fit specific requirements, such as creating a more horizontal image for a wallpaper or a vertical one for mobile phone screens.

  • What does the 'CFG' setting represent and how does it affect the generated images?

    -The 'CFG' setting represents how literally the AI will follow the user's prompt. A default setting of seven provides a balance between following the prompt and allowing for creative, unexpected results. Setting it to zero may produce unrelated images, while setting it to the maximum results in images that closely adhere to the prompt but may be less experimental.

  • How does the 'steps' setting influence the image generation process?

    -The 'steps' setting determines how much extra time the AI spends diffusing the image. A lower setting results in faster image completion but may appear less sophisticated, while a higher setting generates more detailed and refined images, albeit at a longer processing time.

  • What is the function of the 'number of images' setting?

    -The 'number of images' setting controls how many images the AI generates each time the user runs the process. The speaker has it set to 9, but users can adjust this number based on their preferences, choosing to generate fewer or more images as desired.

  • What does the speaker admit about not understanding regarding the 'sampler' setting?

    -The speaker admits that they do not know what the 'sampler' setting does or how it affects the results. They mention different samplers like 'klms', 'kdpm2', 'ancestral kdpm2', 'cooler', 'ancestral cooler', 'plms', and 'ddim', but acknowledge that they have not yet figured out their specific effects.

  • How does the image editor feature work in the stable diffusion site?

    -The image editor feature allows users to upload any image and then scale, pan, erase, or restore parts of it. The brush size, sharpness, and strength can be adjusted, as well as the image opacity, which controls the transparency of the entire image.

  • What issue does the speaker mention with the image editor when using Firefox?

    -The speaker notes that there is a glitch with the image editor when using Firefox, where the tools do not appear. This issue is specific to Firefox users, and the speaker hopes it will be fixed soon.

  • How can users download the generated images?

    -Users can download all the generated images individually or choose to download them as a single zip file. This feature provides flexibility in how users save and organize their AI-generated images.

  • What technique does the speaker suggest for mutating an image?

    -The speaker suggests using the 'image opacity' setting to mutate an image. By adjusting the transparency, users can control the aggressiveness of the mutation, with lower opacities leading to more significant changes and higher opacities resulting in more subtle alterations.



🌟 Introduction to Stable Diffusion AI Generator

The paragraph introduces the use of the official stable diffusion site for AI image generation. The speaker expresses support for the open-source AI generator and its developers, explaining that purchasing credits on the site directly funds product improvements. Accessibility is emphasized as a reason for using the official website, catering to users who may not have the technical expertise or resources to install and run the software locally. The paragraph also mentions that while the service is paid, alternatives exist, though they may be slower.

🎨 Customizing Image Generation with Stable Diffusion

This section delves into the features of the stable diffusion site, highlighting the weapon height slider controller for adjusting image dimensions, the CFG setting for how closely the AI follows the user's prompt, and the steps setting that affects image generation time and quality. The speaker explains the trade-offs between speed and sophistication in image generation and touches on the number of images setting, which determines how many variations are produced per prompt.

🖌️ Exploring Sampler Settings and Download Options

The speaker admits uncertainty about the sampler settings but encourages users to experiment with them. The paragraph discusses the download options available for generated images, including the ability to download all images at once or as a zip file. The paragraph also introduces the image editor, which allows users to upload and modify images, although it notes a glitch with the tools not appearing in Firefox and another issue with the brush tool disabling when the mouse goes outside the canvas.

🌀 Image Mutation and Editing Techniques

The final part of the paragraph focuses on image mutation using image opacity and the various editing tools available, such as scaling, panning, erasing, and restoring the original image. The speaker provides insights on brush size, sharpness, and opacity control, as well as the functionality of the strength and image opacity settings. The paragraph concludes with a brief mention of changing the width and height of the image and a prompt for users to share their experiences with the image editor.



💡stable diffusion site

The term 'stable diffusion site' refers to the official website used for generating AI-based images. It is the platform where users can access and utilize the AI generator for their creative needs. In the context of the video, the speaker prefers this site due to their support for the developers and its accessibility for the average user, highlighting its importance in facilitating AI-generated content creation for a broader audience.

💡open source AI generator

An 'open source AI generator' is a software tool that is publicly accessible and allows users to view and modify its source code. The video emphasizes the speaker's preference for the stable diffusion site over other AI generators, not because they are unaware of open source options, but due to their desire to support the developers directly and keep the series accessible.

💡buy credits

The term 'buy credits' refers to the process of purchasing usage rights or access to additional features on the AI generator platform. The video explains that buying credits on the stable diffusion site directly supports the developers, enabling them to enhance and maintain the product, which is a key aspect of the speaker's decision to use this particular site.


CFG, or Configuration, is a parameter within the AI generator that determines the strictness with which the AI follows the user's prompt. A higher CFG value leads to more literal interpretations of the prompt, while a lower value allows for more abstract or unrelated images. The video describes the importance of balancing CFG to achieve the desired level of creativity and adherence to the prompt in the generated images.


In the context of the AI generator, 'steps' refers to the computational process of refining the image generation. The more steps taken, the more sophisticated and detailed the final image becomes. However, increasing the number of steps also increases the time required to generate the image. The video discusses the trade-off between speed and quality when adjusting the number of steps.

💡number of images

The 'number of images' setting determines how many different AI-generated images the user will receive per prompt. The video mentions that the user can customize this setting to receive multiple interpretations of their prompt, offering a range of creative options to choose from.


A 'sampler' in the context of the AI generator is a method or algorithm used to select or generate elements within the image creation process. While the speaker admits to not fully understanding its function, they mention it as one of the settings available to users, suggesting that it may influence the style or quality of the generated images.

💡image editor

The 'image editor' is a feature within the stable diffusion site that allows users to upload and modify existing images. The video describes the functionalities of the image editor, such as scaling, panning, erasing, and restoring parts of the image, providing users with tools to further refine their AI-generated content according to their preferences.


In the context of the video, 'mutation' refers to the process of altering an existing image to create a new variation. By adjusting the image opacity, the AI attempts to generate a slightly modified version of the original image, introducing subtle changes and allowing for experimentation with the final output.


The term 'accessibility' in the video refers to the ease with which users can utilize the AI generator. The speaker emphasizes the importance of choosing a platform that is accessible to the average person, regardless of their technical expertise, in order to make the technology more inclusive and widely usable.


The speaker expresses support for the official stable diffusion site and its AI generator.

The AI generator's development is funded by users purchasing credits, which directly benefits the developers.

The official site is used to keep the series accessible to the average person, even though the software can be installed locally.

The site offers a streamlined interface with a default dark theme.

The weapon height slider controller allows users to change the dimensions of the image based on their needs.

CFG setting determines how closely the AI follows the user's prompt, with a default value of seven.

Higher CFG values result in more accurate but less experimental image results.

The steps setting controls how much extra time is spent on generating the image, affecting its quality.

The number of images setting determines how many images are generated per prompt.

Sampler settings affect the image generation process, though their exact function is not clearly understood by the speaker.

Images can be downloaded individually or as a zip file.

The image editor allows users to upload and modify images, with tools like scaling, panning, erasing, and restoring.

The brush tool in the image editor can be adjusted for size, sharpness, and opacity.

A glitch is mentioned where the brush tool becomes disabled if the mouse goes outside the canvas.

The image opacity setting can be used to mutate an image, with more transparency leading to more aggressive mutations.

The speaker encourages users to experiment with the settings to achieve desired results.

The transcript concludes with a positive message, wishing the audience a fantastic day.