Skip to main content

​
What are Models?

Models are the AI engines that power text and image generation inside Canvasit.
They determine how your inputs (such as text prompts, references, or block connections) are interpreted and transformed into creative outputs. There are two main types of models you’ll use:
  • Text Models:
    These models are optimized for generating or transforming text.
    They help refine prompts, expand ideas, or provide descriptive content that can then be used by Image Blocks.
  • Image Models:
    These models specialize in creating visual outputs.
    They take text prompts or reference images as input and generate new images based on your creative direction.

​
Why Models are useful

By combining text and image models, you can:
  • Start with a simple idea written in a Text Block.
  • Use a text model to improve, expand, or stylize that idea.
  • Feed the result into an image model to generate a visual draft.
  • Refine the process iteratively by adjusting prompts or adding Reference Blocks.
This modular approach gives you flexibility: you can experiment with different models to see which best fits your style, project, or use case.
Next, explore Text Models to learn how to create and optimize prompts effectively.