ComfyUI-Florence2 - AI Vision Tools Tool

Overview

ComfyUI-Florence2 is a GitHub repository that integrates Microsoft's Florence-2 vision foundation model into ComfyUI. It enables prompt-based vision and vision-language tasks such as captioning, object detection, segmentation, and Document Visual Question Answering on scanned documents.

Key Features

• Integrates Microsoft Florence-2 into the ComfyUI framework
• Prompt-based vision and vision-language task support
• Image captioning capabilities
• Object detection support
• Image segmentation support
• Document Visual Question Answering (DocVQA) for scanned documents
• Designed to run within ComfyUI node-based workflows

Ideal Use Cases

• Generate image captions from photographic inputs
• Detect and localize objects in images
• Perform segmentation for image analysis
• Answer questions about scanned documents (DocVQA)
• Prototype vision-language workflows inside ComfyUI

Getting Started

• Clone the ComfyUI-Florence2 repository from GitHub
• Install or open your ComfyUI environment
• Read the repository README for integration instructions
• Place Florence-2 model files as instructed by the repo
• Load the Florence-2 integration nodes inside ComfyUI
• Run example prompts for captioning or DocVQA

Pricing

Pricing not disclosed; repository is available on GitHub.

Limitations

• Requires a working ComfyUI environment
• Repository-based integration requires technical installation and configuration

Key Information

• Category: Vision Tools
• Type: AI Vision Tools Tool