sciver

MCPOfficialPaddlePaddle

PaddleOCR MCP

Recognizes text and parses documents with tables, formulas and handwriting.

The PaddleOCR project’s server exposes its OCR and document parsing pipelines: it extracts text, tables, formulas and handwriting from PDFs and images into an editable form. It runs locally (the [local-cpu] extra also installs the PaddlePaddle engine) or through an API: the official one on Baidu AI Studio (access token), Qianfan or your own PaddleOCR server.

Installation

Manual install

$pip install -U paddleocr-mcp

Details

Works in
Claude, VS Code
Transport
local
License
Apache-2.0
Tasks
Docs, Research
Rating
Source
github.com/PaddlePaddle/PaddleOCR/tree/main/mcp_server (opens in a new tab)

Tags

  • ocr
  • pdf
  • tables
  • handwriting
  • paddleocr

Log in to rate

A pack is your own set of skills and MCP servers. sciver loads it in your AI with one command.

View in the library