jkawamoto avatar

Mcp Florence2

jkawamoto/mcp-florence2

Connects Claude to Microsoft's Florence-2 vision model for local image processing. Exposes two tools: ocr for extracting text from images and PDFs, and caption for generating descriptive summaries of image content. Works with both local files and web URLs. Reach for this when you need offline OCR capabilities or want to generate alt text and image descriptions without hitting external APIs. Ships as a pre-built bundle that drops right into Claude Desktop, or runs via uvx for Goose and LM Studio. The model downloads locally on first run, so you own the processing pipeline end to end.

GitHub stars
7
tools
3
transport
STDIO
last updated
Oct 28, 2026

Tools

3 tools

Public tool metadata for what this MCP can expose to an agent.

ocrExtracts all visible text from images or multi-page PDF documents.

Extracts all visible text from images or multi-page PDF documents.

No parameter schema in public metadata yet.

captionGenerates comprehensive, detailed captions describing what is depicted in the image.

Generates comprehensive, detailed captions describing what is depicted in the image.

No parameter schema in public metadata yet.

processExecutes custom Florence-2 task prompts directly against the input image or PDF for advanced computer vision workflows (such as object detection or region captioning).

Executes custom Florence-2 task prompts directly against the input image or PDF for advanced computer vision workflows (such as object detection or region captioning).

No parameter schema in public metadata yet.

Sponsors

Readme

English | 日本語

Florence-2 MCP Server

uv Python Version Python Application pre-commit GitHub License

MseeP.ai Security Assessment Badge


View the full README on GitHub

Details

Repository
github.com/jkawamoto/mcp-florence2
Package
https://github.com/jkawamoto/mcp-florence2/releases/download/v0.5.3/mcp-florence2.mcpb
Transport
STDIO
Registry
MCP registry, active

More from jkawamoto