CLIPTextEncodeFlux - ComfyUI Built-in Node Documentation - ComfyUI

CLIPTextEncodeFlux is an advanced text encoding node designed for the Flux architecture. It processes two separate text inputs through different encoders—CLIP-L and T5XXL—and combines them with a guidance scale to produce a unified conditioning output for image generation.

Inputs

Parameter	Description	Data Type	Required	Range
`clip`	A CLIP model that supports the Flux architecture, including both CLIP-L and T5XXL encoders.	CLIP	Yes	-
`clip_l`	Text input processed by the CLIP-L encoder. Suitable for concise keyword descriptions, such as style or theme. Supports multiline input and dynamic prompts.	STRING	Yes	-
`t5xxl`	Text input processed by the T5XXL encoder. Suitable for detailed natural language descriptions, expressing complex scenes and details. Supports multiline input and dynamic prompts.	STRING	Yes	-
`guidance`	Controls the influence of text conditions on the generation process. Higher values mean stricter adherence to the text. Default: 3.5. Adjustable in increments of 0.1.	FLOAT	Yes	0.0 - 100.0

Outputs

Output Name	Description	Data Type
`CONDITIONING`	Contains the fused embeddings from both encoders and the guidance parameter, used for conditional image generation.	CONDITIONING

This documentation was AI-generated. If you find any errors or have suggestions for improvement, please feel free to contribute! Edit on GitHub

Source fingerprint (SHA-256): 022928fa6917102f5dc599364df9541b2451b42eb36a11813931b5fd71990b74