CLIP Text Encode SDXL Refiner

CLIPTextEncodeSDXLRefiner Documentation

• Class name: CLIPTextEncodeSDXLRefiner

• Category: advanced/conditioning

• Output node: False

This node specializes in refining the encoding of text inputs using CLIP models, enhancing the conditioning for generative tasks by incorporating aesthetic scores and dimensions.

CLIPTextEncodeSDXLRefiner Input types

Parameter	Comfy dtype	Description
clip	CLIP	A CLIP model instance used for text tokenization and encoding, central to generating the conditioning.
ascore	FLOAT	The aesthetic score parameter influences the conditioning output by providing a measure of aesthetic quality.
width	INT	Specifies the width of the output conditioning, affecting the dimensions of the generated content.
height	INT	Determines the height of the output conditioning, influencing the dimensions of the generated content.
text	STRING	The text input to be encoded, serving as the primary content descriptor for conditioning.

CLIPTextEncodeSDXLRefiner Output types

Parameter	Comfy dtype	Description
conditioning	CONDITIONING	The refined conditioning output, enriched with aesthetic scores and dimensions for enhanced content generation.