CLIPTextEncode - ComfyUI Built-in Node Documentation

Complete documentation for the CLIPTextEncode node in ComfyUI. Learn its inputs, outputs, parameters and usage.

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

CLIP Text Encode (CLIPTextEncode) acts as a translator, converting your text descriptions into a format that AI can understand. This helps the AI interpret your input and generate the desired image.

Think of it as communicating with an artist who speaks a different language. The CLIP model, trained on vast image-text pairs, bridges this gap by converting your descriptions into “instructions” that the AI model can follow.

Inputs

Parameter	Data Type	Required	Range	Description
text	STRING	Yes	Any text	The text to be encoded. Supports multi-line input and dynamic prompts.
clip	CLIP	Yes	Loaded CLIP models	The CLIP model used for encoding the text.

Outputs

Output Name	Data Type	Description
CONDITIONING	CONDITIONING	A conditioning containing the embedded text used to guide the diffusion model.

Prompt Features

Embedding Models

Embedding models allow you to apply specific artistic effects or styles. Supported formats include .safetensors, .pt, and .bin. To use an embedding model:

1. Place the file in the ComfyUI/models/embeddings folder.
2. Reference it in your text using embedding:model_name.

Example: If you have a model named EasyNegative.pt in your ComfyUI/models/embeddings folder, then you can use it like this:

worst quality, embedding:EasyNegative, bad quality

IMPORTANT: When using embedding models, verify that the file name matches and is compatible with your model’s architecture. For example an embedding designed for SD1.5 will not work correctly for a SDXL model.

Prompt Weight Adjustment

You can adjust the importance of certain parts of your description using parentheses. For example:

• (beautiful:1.2) increases the weight of “beautiful”.
• (beautiful:0.8) decrease the weight of “beautiful”.
• Plain parentheses (beautiful) will apply a default weight of 1.1.

You can use the keyboard shortcuts ctrl + up/down arrow to quickly adjust weights. The weight adjustment step size can be modified in the settings.

If you want to include literal parentheses in your prompt without changing the weight, you can escape them using a backslash e.g. \(word\).

Wildcard/Dynamic Prompts

Use {} to create dynamic prompts. For example, {day|night|morning} will randomly select one option each time the prompt is processed.

If you want to include literal curly braces in your prompt without triggering dynamic behavior, you can escape them using a backslash e.g. \{word\}.

Source fingerprint (SHA-256): e8f286cdec879c529270e110ccf5959ed6df77737cfb5a8019379afac9266118