Visit ComfyUI Online for ready-to-use ComfyUI environment
Converts images to text prompts and tags using advanced AI models for AI artists, offering full/simplified prompts and tag sets with scoring and pattern removal options.
The Image2TextWithTags
node is designed to convert images into descriptive text prompts and tags, enhancing the understanding and categorization of visual content. This node leverages advanced AI models to generate detailed and accurate descriptions, making it a valuable tool for AI artists who need to annotate images or create prompts for further processing. The node can produce a full prompt, a simplified prompt, and a set of tags, providing a comprehensive textual representation of the image. Additionally, it offers options to include a scoring mechanism and to remove specific patterns from the generated tags, ensuring the output is tailored to your needs.
This parameter specifies the AI model to be used for generating the text and tags from the image. The model's performance and the quality of the output depend on the chosen model's capabilities.
This parameter takes the input image that you want to convert into text and tags. The image should be in a format supported by the node, such as JPEG or PNG.
This parameter allows you to input a specific query or context that guides the AI model in generating the text and tags. It helps in tailoring the output to be more relevant to your needs.
This parameter provides an additional layer of customization by allowing you to input a custom query that further refines the generated text and tags. It works in conjunction with the main query to produce more precise results.
This boolean parameter, when set to True, enables logging of the process, providing insights into the node's execution. This can be useful for debugging or understanding how the node generates its output. The default value is False.
This boolean parameter, when enabled, includes a scoring mechanism in the output, which can help in evaluating the relevance or quality of the generated text and tags. The default value is False.
This boolean parameter, when set to True, removes specific patterns, such as "1girl," from the generated tags. This is useful for filtering out unwanted or irrelevant tags from the output. The default value is True.
This output parameter provides a comprehensive textual description of the input image, combining both the main query and the custom query results. It offers a detailed and holistic view of the image content.
This output parameter delivers a simplified version of the textual description, focusing on the main elements of the image. It is useful for quick reference or when a concise description is needed.
This output parameter generates a set of tags that capture the key elements and attributes of the image. These tags are prioritized by relevance and are useful for categorization, search, and further processing of the image.
score
parameter if you need to evaluate the quality or relevance of the generated text and tags.remove_1girl
parameter to filter out unwanted tags, ensuring the output is more relevant to your needs.ยฉ Copyright 2024 RunComfy. All Rights Reserved.