CivitAI Tutorial

AI in 5 Minutes
19 Dec 202303:06

TLDRIn this tutorial video, the host guides viewers through the process of using CivitAI, a generative AI platform. After opening a browser and navigating to the CivitAI website, users are instructed to create an account and select a model, such as 'epic realism', from the available options. The tutorial then covers how to input prompts, adjust settings like CFG scale and sampler, and customize steps for image generation. The host also demonstrates how to copy a prompt from chat GPT and generate four images using the platform. Additionally, viewers learn how to access the queue window to view past generations and their details, including negative prompts and the base model used. The video concludes with a reminder to generate their own images and enjoy the process, encouraging viewers to create unique and captivating visuals.


  • 🌐 Open your web browser and navigate to by typing it in the address bar and hitting enter.
  • πŸ“ Create an account on the website if you haven't already.
  • πŸ”‘ Click on the 'Create' button to start the image generation process.
  • πŸ–ΌοΈ Select a model for your AI image generation, such as 'Epic Realism', from the options provided.
  • ✍️ Enter your desired prompt in the designated field, and specify any negative prompts or mature content preferences.
  • πŸ“ Choose your aspect ratio for the generated image.
  • πŸŽ›οΈ Adjust the CFG scale and select a sampler for the image generation process.
  • πŸ”„ Customize the number of steps for the generation process if needed.
  • 🧠 Leave the SE option to 'random' for a more varied outcome.
  • πŸ”„ Adjust the clip skip and add a VAE if necessary, and set the quantity of images you want to generate.
  • πŸ–₯️ Click 'Generate' to start creating your images based on the provided prompt.
  • πŸ“ Copy and paste prompts from external sources like chat GPT directly into the prompt field for generation.
  • πŸ—οΈ Use the queue window to view past generations, including details like prompts and settings used.
  • ↩️ Regenerating images is as simple as selecting a previous generation and clicking 'Generate' again.
  • πŸ–ΌοΈ Access all your generated images individually through the 'Feed' button.

Q & A

  • What is the first step to access the AI platform mentioned in the tutorial?

    -The first step is to open your browser, type in 'sa AI' in the address bar, and hit enter to navigate to the website.

  • How do you create an account on the AI platform?

    -Once you've accessed the website, you need to create an account by following the site's registration process.

  • What is the 'create button' mentioned in the script?

    -The 'create button' is likely a button on the AI platform's interface that users click to start the process of generating an AI model.

  • How can you change the model on the AI platform?

    -You can change the model by clicking on a button on the left side of the window, which allows you to select from different available models.

  • What is the purpose of the 'prompt' and 'negative prompt' in the AI image generation process?

    -The 'prompt' is a text input that guides the AI to generate an image according to the user's description. The 'negative prompt' is used to specify what should not be included in the generated image.

  • What does 'NSFW' stand for and where is it relevant in the script?

    -NSFW stands for 'Not Safe For Work'. In the script, it is relevant when the user wants to create mature or adult content, which is indicated by a selection in the platform's interface.

  • What is the 'CFG scale' and how can it be adjusted?

    -The 'CFG scale' likely refers to a configuration setting related to the image generation process. It can be adjusted within the platform's interface to control certain aspects of the generated image.

  • What does the 'SE option' represent and what is the suggestion for it?

    -The 'SE option' is not explicitly defined in the script, but it likely stands for 'Style Embedding' or a similar feature. The suggestion is to leave it random.

  • How can you customize the number of steps in the image generation process?

    -You can customize the number of steps by scrolling down to the relevant section in the interface and adjusting the settings as needed.

  • What is the 'queue window' and how does it function?

    -The 'queue window' is a feature that displays all the past generations of images created by the user, along with their prompts and necessary details for each generation.

  • How can you regenerate an image using the same settings?

    -To regenerate an image, locate the specific image or settings in the queue window, click on the corresponding window, and then click on 'generate' to recreate the image.

  • What is the 'feed button' and what does it display?

    -The 'feed button' likely displays all the images that have been generated by the user, showing them one by one for review or sharing.



🎨 Introduction to Creating with SAI

The speaker begins by introducing the tutorial, which is about creating images using SAI (Stable Diffusion AI). They instruct viewers to open a web browser, navigate to the SAI website, and create an account. After account creation, they guide viewers to the 'create' button and explain the process of selecting a model, such as 'epic realism,' and customizing settings like the prompt, negative prompt, and aspect ratio. The speaker also details how to adjust the CFG scale, sampler, steps, and other options like the SE option and noun. They demonstrate how to generate images using a prompt from chat GPT and how to view and regenerate images from the queue window. The tutorial concludes with a mention of the feed button, which shows all generated images individually.




A browser is a software application for accessing information on the World Wide Web. In the context of the video, the browser is the tool used to navigate to the AI platform's website, which is the first step in the tutorial. It's essential for users to understand how to use a browser to access online services and applications.


An account, in the context of the video, refers to a user profile created on the AI platform. This account is necessary for personalized settings and to save the user's generated content. The video emphasizes the importance of creating an account to engage with the AI service effectively.

πŸ’‘create button

The 'create button' is a user interface element within the AI platform that initiates the process of generating content. In the video, clicking this button leads users to the content generation interface, where they can customize their AI-generated images. It is a crucial step in the tutorial, as it marks the transition from setup to content creation.


In the context of the video, a 'model' refers to the underlying AI algorithms that are used to generate the images. The user can select different models, such as 'epic realism,' which influences the style and quality of the generated content. The choice of model directly affects the outcome, making it an important concept in the tutorial.


A 'prompt' is the input text or idea provided by the user that the AI uses as a basis to generate content. In the video, the user is instructed to type in their prompt, which guides the AI in creating the desired images. The prompt is a key element in the content generation process, as it sets the theme and direction for the AI's output.

πŸ’‘negative prompt

A 'negative prompt' is a set of instructions or preferences specified by the user to exclude certain elements from the AI-generated content. In the video, it is mentioned as a feature to prevent mature or NSFW (Not Safe For Work) content from being generated. This ensures that the AI's output aligns with the user's preferences and content guidelines.

πŸ’‘CFG scale

CFG scale refers to the 'Control Flow Graph' scale, which is a parameter in the AI platform that adjusts the level of detail and complexity in the generated images. In the tutorial, the user is given the option to adjust the CFG scale to achieve their desired level of intricacy in the AI-generated content. This feature is important for users looking to fine-tune the quality of their outputs.


A 'sampler' in the context of the AI platform is a tool or algorithm that selects random elements from a given set to contribute to the final output. The video mentions changing the sampler, which can affect the diversity and uniqueness of the generated images. It is a key concept for users who want to experiment with different styles and outcomes in their AI-generated content.


In the context of the video, 'steps' refer to the number of iterations or stages the AI goes through to generate the final image. Customizing the steps can influence the refinement and polish of the AI's output. The tutorial suggests leaving it at default, but it's an option for users who want more control over the generation process.

πŸ’‘SE option

The 'SE option' stands for 'Style Embossment,' which is a feature in the AI platform that emphasizes the style of the input prompt in the generated images. The video suggests leaving it at 'random,' but this option allows users to control the stylistic influence of the input on the output, adding another layer of customization to the content generation process.


VAE stands for 'Variational Autoencoder,' which is a type of AI model used for generating images. In the video, the user can adjust the quantity of images generated by the VAE. This option provides users with control over the amount of content they wish to produce, allowing for flexibility in their creative process.

πŸ’‘queue window

The 'queue window' is a part of the AI platform's interface that displays all the previous generations of images, along with their associated prompts and settings. In the video, it is mentioned as a way to review and regenerate past content. This feature is useful for users who want to revisit and refine their previous creations.


The tutorial begins with opening a browser and navigating to the AI website.

Creating an account is necessary to use the AI service.

The 'create' button is used to start the AI image generation process.

Models can be selected and changed on the left side of the window.

The 'epic realism' model is chosen as an example for image generation.

Prompts and negative prompts are typed into designated fields.

NSFW content can be created by selecting the appropriate option.

The aspect ratio for the generated images can be adjusted.

CFG scale and sampler settings can be modified for different outcomes.

Steps can be customized for the image generation process.

The SE option is suggested to be left random.

The clip skip and the quantity of generated images can be adjusted.

The 'generate' button initiates the image creation.

A prompt from chat GPT is used to demonstrate the process.

Four images are generated as a result of the prompt.

The queue window displays all past generations with their details.

Additional details such as negative prompts, sampler, steps, and model are provided for each generation.

Images can be regenerated using the provided window and button.

The feed button shows all generated images one by one.

The tutorial concludes with a suggestion to generate similar images and enjoy the process.