ComfyUI > Nodes > ComfyUI Janus Pro Vision > Janus Vision 7b Pro (Chat)

ComfyUI Node: Janus Vision 7b Pro (Chat)

Class Name

UnifiedVisionAnalyzer

Category
JanusVision
Author
ShmuelRonen (Account age: 1462days)
Extension
ComfyUI Janus Pro Vision
Latest Updated
2025-03-05
Github Stars
0.02K

How to Install ComfyUI Janus Pro Vision

Install this extension via the ComfyUI Manager by searching for ComfyUI Janus Pro Vision
  • 1. Click the Manager button in the main menu
  • 2. Select Custom Nodes Manager button
  • 3. Enter ComfyUI Janus Pro Vision in the search bar
After installation, click the Restart button to restart ComfyUI. Then, manually refresh your browser to clear the cache and access the updated list of nodes.

Visit ComfyUI Online for ready-to-use ComfyUI environment

  • Free trial available
  • 16GB VRAM to 80GB VRAM GPU machines
  • 400+ preloaded models/nodes
  • Freedom to upload custom models/nodes
  • 200+ ready-to-run workflows
  • 100% private workspace with up to 200GB storage
  • Dedicated Support

Run ComfyUI Online

Janus Vision 7b Pro (Chat) Description

Versatile image analysis node with chat interaction, part of JanusVision suite, leveraging advanced vision models for detailed insights.

Janus Vision 7b Pro (Chat):

The UnifiedVisionAnalyzer is a versatile node designed to perform comprehensive image analysis with the added capability of engaging in chat-based interactions. This node is part of the JanusVision suite and is tailored to provide detailed insights into images by leveraging advanced vision models. It allows you to input an image and receive a descriptive analysis based on a given prompt. Additionally, it supports a chat mode that enables interactive discussions about the image, making it a powerful tool for AI artists who wish to explore and understand visual content more deeply. The node is equipped with various configurable parameters that allow you to fine-tune the analysis process, ensuring that the results are tailored to your specific needs. Whether you are looking to generate detailed image descriptions or engage in a dynamic conversation about visual content, the UnifiedVisionAnalyzer offers a robust solution.

Janus Vision 7b Pro (Chat) Input Parameters:

janus_model

This parameter specifies the vision model to be used for image analysis. It is crucial as it determines the underlying capabilities and performance of the analysis. The model should be compatible with the JanusVision framework.

image_a

This is the primary image input for analysis. The node will process this image to generate a descriptive response based on the provided prompt. The quality and content of this image directly impact the analysis results.

prompt

The prompt is a string input that guides the analysis process. It can be a question or a statement that you want the node to address regarding the image. The prompt supports multiline input and defaults to "Please describe this image." This flexibility allows for tailored and specific inquiries about the visual content.

chat_mode

A boolean parameter that enables or disables the chat functionality. When set to true, the node engages in a conversational mode, allowing for interactive discussions about the image. The default value is false.

seed

An integer value used to initialize the random number generator, ensuring reproducibility of results. The default is 42, with a range from 0 to 18,446,744,073,709,551,615.

temperature

A float parameter that controls the randomness of the response generation. Lower values make the output more deterministic, while higher values introduce more variability. The default is 0.1, with a range from 0.0 to 2.0.

top_p

This float parameter is used for nucleus sampling, determining the cumulative probability threshold for token selection. It helps balance creativity and coherence in the generated responses. The default is 0.95, with a range from 0.0 to 1.0.

max_tokens

An integer that sets the maximum number of tokens for the generated response. This limits the length of the output, with a default of 512 and a range from 1 to 2048.

image_size

Specifies the size of the image to be analyzed, in pixels. This parameter affects the resolution and detail level of the analysis. The default is 1024, with a range from 512 to 2048.

frame_size

An integer that defines the size of the frame used in the analysis process. It influences the granularity of the image processing. The default is 2, with a range from 1 to 10.

reset_chat

A boolean parameter that, when set to true, clears the chat history, allowing for a fresh start in the conversation. The default value is false.

image_b

An optional secondary image input that can be used for comparative analysis or additional context. This parameter is not required but can enhance the depth of the analysis.

Janus Vision 7b Pro (Chat) Output Parameters:

response

The response is a string output that provides a detailed analysis or description of the input image based on the given prompt. It reflects the node's interpretation and understanding of the visual content, offering insights and information that align with the prompt's intent.

chat_history

This output is a string that contains the history of the chat interactions if chat mode is enabled. It provides a record of the conversation, allowing you to review the dialogue and understand the progression of the discussion about the image.

Janus Vision 7b Pro (Chat) Usage Tips:

  • To achieve more creative and varied responses, consider adjusting the temperature and top_p parameters. Higher values can lead to more diverse outputs.
  • Use the reset_chat parameter to clear the chat history when starting a new analysis session, ensuring that previous interactions do not influence the current analysis.
  • Experiment with different prompts to explore various aspects of the image. The prompt can significantly influence the focus and detail of the analysis.

Janus Vision 7b Pro (Chat) Common Errors and Solutions:

"Invalid model specified"

  • Explanation: The janus_model parameter is not set to a compatible model.
  • Solution: Ensure that the model specified is compatible with the JanusVision framework and correctly loaded.

"Image input is missing"

  • Explanation: The image_a parameter is not provided, which is essential for analysis.
  • Solution: Provide a valid image file for the image_a parameter to enable the analysis process.

"Prompt is too long"

  • Explanation: The input prompt exceeds the maximum allowed length.
  • Solution: Shorten the prompt to fit within the acceptable length, ensuring it remains clear and concise.

"Exceeded max tokens limit"

  • Explanation: The generated response exceeds the max_tokens limit.
  • Solution: Increase the max_tokens parameter if a longer response is needed, or refine the prompt to focus the analysis.

Janus Vision 7b Pro (Chat) Related Nodes

Go back to the extension to check out more related nodes.
ComfyUI Janus Pro Vision
RunComfy
Copyright 2025 RunComfy. All Rights Reserved.

RunComfy is the premier ComfyUI platform, offering ComfyUI online environment and services, along with ComfyUI workflows featuring stunning visuals. RunComfy also provides AI Playground, enabling artists to harness the latest AI tools to create incredible art.