AI Image Generation Program - User Manual
This manual will guide you on how to use the AI Image Generation program. This program allows you to create images from text prompts, modify the output using art styles and system directives, and manage your generated images.
I. Program Overview:
The program interface is divided into two main columns:
Left Column (Parameters): This column contains all the settings and controls for generating images. You'll input your desired image description, select art styles, choose models, and configure various other parameters here.
Right Column (Image Display): This column displays the images generated by the program. New images appear at the top.
II. Getting Started:
Accessing the Program: The program is a web-based application. Open it in your web browser (e.g., Chrome, Firefox, Safari).
Initial Load: When the program loads:
Dark Mode will be enabled by default.
The "Enhance" button in parameters is enabled.
The system will select some default system directives.
III. Key Components and their Functions:
A. Top Controls:
"AI Image Generation Parameters" Heading: Simply a title for the left column.
"Dark Mode" Checkbox:
What: Toggles between dark and light mode for the interface.
How: Check the box to enable dark mode (recommended for reduced eye strain). Uncheck to use the light mode.
"Generate Image" Button:
What: Initiates the image generation process using the current settings.
How: Click this button after you have configured all the parameters to your liking. The progress will be displayed below this button.
"Progress: 0%" Display:
What: Shows the progress of the image generation. Once complete, the generated image will appear in the right column.
B. Prompt Input and Manipulation:
"User Prompt" Textarea:
What: This is where you enter a text description of the image you want to generate. The better the prompt, the better the image.
How: Type your desired image description into the text area. Be as descriptive as possible. For example, "A futuristic cityscape at sunset with flying cars."
"Art Style" Selection (Multiple Select):
What: Allows you to select one or more art styles to influence the image generation. Each style has a weight associated with it (e.g., "Art Deco:1.8"). Higher weight values mean the style will have a stronger influence.
How: Click on the art styles you want to apply. You can select multiple styles by holding down Ctrl (Windows) or Command (Mac) while clicking.
"Enhance" Button:
What: Uses an AI text model to rewrite and improve the user prompt, generating a more detailed and effective description for image generation.
How:
Write your initial prompt in the "User Prompt" box.
Click the "Enhance" button.
The enhanced prompt will appear in the "Enhanced Prompt" textarea.
"Clear" Button:
What: Clears all generated images from the display in the right column.
How: Click the "Clear" button to remove the currently displayed images.
"Down" Button: (Download All Images)
What: Downloads all generated images as a single ZIP file. Requires the JSZip library to be loaded.
How: Click the "Down" button to start the download.
"Seed" Button:
What: Generates a random seed number. The seed is used to control the randomness of the image generation process. Using the same prompt and seed will produce the same image.
How: Click the "Seed" button. A random number will be placed in the "Seed" field. To recreate the same image, keep the same seed and the same prompt.
"Save" Button:
What: This button allows you to save the current settings (User Prompt, Enhanced Prompt, System Prompt, Text Model, and Image Model). This is useful for keeping track of settings that produce desirable results.
How: Click the "Save" button after you have configured your settings.
"System Directive" Button:
What: Opens a modal (pop-up window) where you can select and modify system directives. System directives influence the behavior of the AI text model used for enhancing prompts. These directives act as instructions given to the AI to control its behavior.
How:
Click the "System Directive" button. A pop-up window will appear.
Select a directive from the dropdown menu. The directive's text will be loaded into the textarea below.
You can modify the text of the directive in the textarea.
Click "Save Directives" to save your changes. (Note: The save function might not be fully implemented in this version.)
Click "Reset to Default" to revert the directive to its original text.
Click "Close" to close the modal.
"Enhanced Prompt" Textarea:
What: This text area displays the enhanced prompt generated by the "Enhance" button. The image will be generated based on this enhanced prompt.
How: This area is read-only. The text is automatically updated when you click the "Enhance" button. You can, however, manually edit the "Enhanced Prompt" if you want more direct control.
"Text Model" Selection:
What: Choose the AI text model that will be used for enhancing the user prompt. Different models have different capabilities and may produce different results.
How: Select a model from the dropdown menu. The options include "openai", "llama", "mistral", and others. The selected model will be used when you click the "Enhance" button. The attributes are described as "modelname|type:chattype|censored:boolean|basemodel:boolean|vision:boolean". Note the model's attribute which determine the output characteristics.
"Image Model" Selection:
What: Choose the AI model that will be used to generate the final image. Different models have different strengths and may produce different results.
How: Select a model from the dropdown menu. The options include "openai", "midijourney", "flux", "turbo" and others. The selected model will be used when you click the "Generate Image" button.
"Seed" Input Field:
What: Allows you to manually enter a seed value. Using the same seed and prompt will generate the same image.
How: Enter a number (e.g., 12345) into the text box. This is useful for reproducing specific images. Click the "Seed" button to automatically generate a random seed.
"Ratio" Selection:
What: Determines the aspect ratio (width to height) of the generated image.
How: Select an aspect ratio from the dropdown menu (e.g., "1:1", "16:9"). The "Height" field will automatically update to reflect the selected ratio.
"Width" Input Field:
What: Sets the width of the generated image in pixels.
How: Enter the desired width in pixels (e.g., 1024). The "Height" field will automatically update if the "Ratio" is set to a specific value.
"Height" Input Field:
What: Shows the calculated height of the generated image based on the selected "Ratio" and "Width."
How: This field is read-only. It automatically updates based on the "Width" and "Ratio" settings.
"Nologo" Checkbox:
What: Determines whether or not the generated image should include a logo.
How: Check the box to remove the logo. Uncheck to include the logo.
"Private" Checkbox:
What: This setting controls the privacy of the generated image.
How: Check the box to make the image private. This is used to tell the image generator API that you want the image to be kept private (though this may depend on the specific API's policies).
"Enhance" Checkbox:
What: This setting is a duplicate and does nothing different.
How: Irrelevant because setting duplicates.
"Safe" Checkbox:
What: This setting enables or disables safety filters.
How: Check the box to enable safety filters. This helps to prevent the generation of NSFW (Not Safe For Work) or offensive images. Uncheck the box to disable the filters. (Use with caution!)
C. Footer (Copyright and Contact Information):
Contains copyright information, contact email, and credits for the AI engine used.
IV. Image Display (Right Column):
Image Container:
What: The area where the generated images are displayed.
How: Images appear here automatically after you click the "Generate Image" button. New images are added to the top.
V. Step-by-Step Image Generation Procedure:
Enter your image description: In the "User Prompt" textarea, type a detailed description of the image you want to create.
Select Art Styles: (Optional) Choose one or more art styles from the "Art Style" selection. Hold Ctrl or Command to select multiple.
Enhance the Prompt: (Recommended) Click the "Enhance" button to improve your prompt using an AI text model.
Select a Text Model: (Optional) Change to a different Text Model for different effects to improve your prompt using an AI text model.
Select an Image Model: Choose the AI model you want to use to generate the image.
Configure Parameters: Adjust the "Ratio", "Width", "Height", "Seed", "Nologo", "Private", and "Safe" settings as desired.
Generate the Image: Click the "Generate Image" button.
View the Result: The generated image will appear in the right column. The "Progress" indicator will show the generation status.
Download the Image: If you want to save the image, right-click on it and choose "Save Image As..."
Download ALL Images: If you want to save multiple images, click the "Down" button.
VI. Advanced Tips:
Prompt Engineering: Experiment with different wording and levels of detail in your prompts to see how they affect the generated images.
Combining Art Styles: Try selecting multiple art styles to create unique and interesting visual effects.
Using Seeds: Use seeds to reproduce specific images. This is useful for experimenting with different settings while keeping the overall image composition the same.
Iterative Refinement: Generate multiple images with slightly different settings to fine-tune the results.
VII. Troubleshooting:
No Image Generated:
Check your prompt for errors or ambiguity.
Try a different image model.
Make sure the "Safe" checkbox is checked (if you're generating SFW images).
Check for error messages in the console (press F12 to open the developer tools and view the console).
Low-Quality Image:
Try enhancing the prompt.
Increase the image width and height.
Select art styles that are known for high detail (e.g., "Hyperrealism").
Try a different image model.
Unwanted Content:
Make sure the "Safe" checkbox is checked.
Refine your prompt to avoid triggering unwanted content.
This manual provides a comprehensive overview of the AI Image Generation program. By following these instructions, you can create stunning and unique images from your text descriptions. Remember to experiment with the different settings and models to discover the full potential of this powerful tool.