Martin Rizzo
ec35c52768
This commit introduces a new Markdown file, `style_prompt_encoder.md`, located in the /docs directory. The document provides detailed information about the functionality and usage of the "Style & Prompt Encoder" node, helping users effectively integrate it into their image generation processes.
2.4 KiB
Style & Prompt Encoder
Encodes the result of fusing a prompt with a visual style, generating embeddings that guide the diffusion process. This node is similar to ComfyUI's native "CLIP Text Encode (Prompt)", but it includes the option to select a visual style for the generated image.
Inputs
clip
The text encoder model used to encode the prompt and style.
customization
Optional input that can remain disconnected. Allows users to configure styles according to their preferences by connecting a multi-line string that redefines one or more available styles. Each style definition starts with ">>>" followed by the name of the style, then includes lines describing the template for that style. The template must include "{$@}" where the user's prompt will be inserted.
Example:
>>>Phone Photo
YOUR CONTEXT:
Your photographs has android phone cam-quality.
Your photographs exhibit surprising compositions, sharp complex backgrounds,
natural lighting, and candid moments that feel immediate and authentic.
Your photographs are actual gritty candid photographic background.
YOUR PHOTO:
{$@}
>>>Casual Photo
YOUR CONTEXT:
You are an amateur documentary photographer taking low quality photos.
Your photographs exhibit sharp backgrounds, unpolished realism with natural lighting,
and candid friendship-level moments that feel immediate and authentic.
YOUR PHOTO:
{$@}
category
The category of the style you want to apply. Changing this option updates the list of available styles in style_to_apply to show only those that match the selected category. Currently, there are 4 categories:
- photo: Contains photographic and realistic styles
- illustration: Includes illustration, drawing, comic, anime, etc.
- other: Styles that don't fit into the previous groups
- custom: Free-form styles for user configuration
style
Selects the desired style to apply or "none" if you prefer not to use any predefined style.
text
This is where you input your prompt. For better style application, avoid including messages in the prompt that might interfere with the selected style.
Outputs
conditioning
The encoded prompt embeddings with applied style, used to guide the model during generation.
string
The prompt text after applying the selected style. Typically not used, but can be employed for advanced purposes.