AI Model Categories & Types
Fater provides a diverse range of Artificial Intelligence (AI) models to assist with various creative tasks. To help you find the right tool for the job, these models are organized into Categories, and often further into Subcategories and Control Types, within the Model Selector (found in the Left Sidebar of the Image, Asset, and Video Editors).
Understanding these classifications will help you navigate the available AI capabilities more effectively.
Main AI Model Categories
The Model Selector groups models into these primary categories, each accessible via a main tab:
Generate (✨ Sparkles Icon):
Purpose: Models in this category are primarily used for creating new visual content from scratch or significantly transforming existing content based on your prompts and control inputs.
Common Uses: Text-to-image generation, generating images based on structural guides (like Edges or Depth maps), or specialized generation tasks like relighting.
Edit (🖌️ Paintbrush Icon):
Purpose: These models are designed for modifying or enhancing existing images, typically within a masked area (inpainting/outpainting).
Common Uses: Removing unwanted objects, seamlessly filling in missing parts of an image, changing specific elements within a scene based on a prompt, or applying stylistic changes to selected regions.
Upscale (🖼️ Image Upscale Icon):
Purpose: Models in this category focus on increasing the resolution and enhancing the detail of existing images.
Common Uses: Improving the quality of lower-resolution images, preparing images for print or high-resolution displays, adding sharpness and clarity.
(Note: Specific models for 3D Asset Generation and Video Generation are selected within their respective Editor Modes and have their own simplified model lists in the Left Sidebar of those modes.)
Subcategories (for Generate & Edit)
Within the "Generate" and "Edit" categories, models are often further divided into subcategories to refine your selection:
Basic:
These models perform their primary function (generation or editing) based mainly on your text prompt and, for "Edit" models, the provided mask and image context. They are a good starting point for general tasks.
Controlled:
These powerful models allow you to guide the generation process with additional visual input in the form of Control Layers (Edges or Depth maps). This provides much greater precision over the structure and composition of the output.
When "Controlled" is selected, an additional "Control Type" dropdown appears:
Contour (Edges): For models that use Edges maps to define outlines and shapes.
Depth: For models that use Depth maps to understand 3D spatial relationships.
Special:
This subcategory groups models with unique functionalities or those that use specific types of input beyond standard prompts or common control layers.
Examples might include models for relighting scenes, specific reference-based editing, or instruction-guided image manipulation.
Model Names
Within the final selection dropdown (after choosing Category, Subcategory, and Control Type if applicable), Fater displays user-friendly names for the models (e.g., "Medium," "Turbo," "High," "Relight"). These names aim to give you an idea of the model's general capability, speed, or quality level within its specific classification.
The AI Model Directory in the Reference section provides more details on each specific user-facing model name and its corresponding parameters.
By navigating these categories and types in the Model Selector, you can pinpoint the most suitable AI model for your creative task, whether you're generating entirely new visuals, precisely editing existing ones, or enhancing image quality.
Last updated