ComfyUI Node: Text to Prompt 🐼

Text to Prompt 🐼:

Text2Prompt is a custom node designed for generating text prompts using a GPT-based model. This node is particularly useful for AI artists who need to create detailed and contextually rich descriptions for their artwork. By leveraging advanced language models, Text2Prompt can generate coherent and contextually appropriate text based on the input parameters provided. This node is ideal for tasks such as generating prompts for stable diffusion models, creating detailed picture descriptions, and enhancing creative writing with AI-generated content. The main goal of Text2Prompt is to simplify the process of generating high-quality text prompts, making it easier for you to focus on your creative work.

Text to Prompt 🐼 Input Parameters:

prompt

This parameter is a string input where you provide the initial text or question that you want the model to generate a response for. It serves as the starting point for the text generation process. The prompt should be clear and concise to ensure the generated text is relevant and coherent. This parameter supports multiline input, allowing you to provide more complex prompts if needed.

model

This parameter specifies the model to be used for text generation. It accepts a TEXT2PROMPT_MODEL type, which is a pre-loaded model instance. The model determines the quality and style of the generated text. By default, it uses the model specified during the node setup.

max_tokens

This integer parameter defines the maximum number of tokens (words or word pieces) that the model can generate in response to the prompt. The default value is 128, but you can adjust it based on your needs. The minimum value is 1, and the maximum value depends on the model's capabilities. Setting this parameter helps control the length of the generated text.

temperature

This float parameter controls the randomness of the text generation process. A lower value (closer to 0) makes the output more deterministic and focused, while a higher value (closer to 1.0) introduces more randomness and creativity. The default value is 0.2, with a range from 0 to 1.0. Adjusting this parameter allows you to fine-tune the balance between coherence and creativity in the generated text.

print_output

This parameter is a dropdown with options "enable" and "disable". When set to "enable", the generated text will be printed to the console, allowing you to see the output directly. The default value is "disable". This is useful for debugging or reviewing the generated text without needing to capture it programmatically.

cached

This parameter is a dropdown with options "YES" and "NO". When set to "YES", the node will use a cached version of the previously generated text if available, which can save time and computational resources. The default value is "NO". This is useful for scenarios where you need consistent outputs for the same input prompt.

prefix

This string parameter allows you to add a prefix to the prompt before it is processed by the model. The default value is "must be in english and describe a picture according to follow the description below within 77 words: ". This can help guide the model to generate text in a specific format or style. The parameter supports multiline input for more complex prefixes.

system_prompt

This parameter provides a selection of predefined system prompts that set the context or role for the model. Options include "You are a helpful assistant.", "你擅长翻译中文到英语。", "你擅长文言文翻译为英语。", "你是绘画大师，擅长描绘画面细节。", and "你是剧作家，擅长创作连续的漫画脚本。". The default value is "You are a helpful assistant.". This helps the model understand the context in which it should generate the text.

Text to Prompt 🐼 Output Parameters:

STRING

The output of the Text2Prompt node is a string containing the generated text. This text is the result of the model processing the input prompt along with any specified parameters such as prefix, system prompt, and temperature. The generated text can be used directly in your projects, providing detailed and contextually appropriate descriptions or prompts for various creative tasks.

Text to Prompt 🐼 Usage Tips:

• Use clear and concise prompts to ensure the generated text is relevant and coherent.
• Adjust the max_tokens parameter to control the length of the generated text, especially for detailed descriptions.
• Experiment with the temperature parameter to find the right balance between coherence and creativity in the generated text.
• Utilize the prefix and system_prompt parameters to guide the model in generating text in a specific format or style.
• Enable print_output for debugging purposes to see the generated text directly in the console.

Text to Prompt 🐼 Common Errors and Solutions:

Error: "Model not found"

• Explanation: This error occurs when the specified model is not available or incorrectly specified.
• Solution: Ensure that the model name is correctly specified and that the model is properly loaded in the system.

Error: "Invalid token length"

• Explanation: This error occurs when the max_tokens parameter is set to a value outside the acceptable range.
• Solution: Adjust the max_tokens parameter to a value within the model's capabilities, typically between 1 and the model's maximum token limit.

Error: "Temperature out of range"

• Explanation: This error occurs when the temperature parameter is set to a value outside the range of 0 to 1.0.
• Solution: Set the temperature parameter to a value between 0 and 1.0 to ensure proper text generation.

Error: "Prompt is too long"

• Explanation: This error occurs when the input prompt exceeds the model's maximum input length.
• Solution: Shorten the input prompt to fit within the model's maximum input length, or split the prompt into smaller segments.