TextGenerate - ComfyUI Built-in Node Documentation

Complete documentation for the TextGenerate 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

The TextGenerate node uses a CLIP model to create text based on a user’s prompt. It can optionally use images, video, or audio as additional context to guide the text generation. You can control the length of the output, enable a thinking mode for supported models, and choose whether to use random sampling with various settings or to generate text without sampling.

Inputs

Parameter	Data Type	Required	Range	Description
clip	CLIP	Yes	N/A	The CLIP model used for tokenizing the prompt and generating text.
prompt	STRING	Yes	N/A	The text prompt that guides the generation. This field supports multiple lines and dynamic prompts. The default value is an empty string.
image	IMAGE	No	N/A	An optional image that can be used alongside the text prompt to influence the generated text.
video	IMAGE	No	N/A	Video frames as an image batch. Assumed to be 24 FPS; subsampled to 1 FPS internally.
audio	AUDIO	No	N/A	An optional audio input that can be used alongside the text prompt to influence the generated text.
max_length	INT	Yes	1 to 2048	The maximum number of tokens the model will generate. The default value is 256.
sampling_mode	COMBO	Yes	"on" "off"	Controls whether random sampling is used during text generation. When set to “on”, additional parameters for controlling the sampling become available. The default is “on”.
thinking	BOOLEAN	No	True or False	Operate in thinking mode if the model supports it. The default value is False.
use_default_template	BOOLEAN	No	True or False	Use the built-in system prompt/template if the model has one. The default value is True. This is an advanced parameter.
temperature	FLOAT	No	0.01 to 2.0	Controls the randomness of the output. Lower values make the output more predictable, higher values make it more creative. This parameter is only available when sampling_mode is “on”. The default value is 0.7.
top_k	INT	No	0 to 1000	Limits the sampling pool to the top K most likely next tokens. A value of 0 disables this filter. This parameter is only available when sampling_mode is “on”. The default value is 64.
top_p	FLOAT	No	0.0 to 1.0	Uses nucleus sampling, limiting choices to tokens whose cumulative probability is less than this value. This parameter is only available when sampling_mode is “on”. The default value is 0.95.
min_p	FLOAT	No	0.0 to 1.0	Sets a minimum probability threshold for tokens to be considered. This parameter is only available when sampling_mode is “on”. The default value is 0.05.
repetition_penalty	FLOAT	No	0.0 to 5.0	Penalizes tokens that have already been generated to reduce repetition. A value of 1.0 applies no penalty. This parameter is only available when sampling_mode is “on”. The default value is 1.05.
presence_penalty	FLOAT	No	0.0 to 5.0	Penalizes new tokens based on whether they have appeared in the text so far, encouraging the model to talk about new topics. This parameter is only available when sampling_mode is “on”. The default value is 0.0.
seed	INT	No	0 to 18446744073709551615	A number used to initialize the random number generator for reproducible results when sampling is “on”. The default value is 0.

Note: The parameters temperature, top_k, top_p, min_p, repetition_penalty, presence_penalty, and seed are only active and visible in the node interface when the sampling_mode is set to “on”.

Outputs

Output Name	Data Type	Description
generated_text	STRING	The text generated by the model based on the input prompt and optional image, video, or audio.

---

Source fingerprint (SHA-256): d2d08577e4d5aeea60f5517377730c4917df607be32f29227b07b1011b0f2c2d