PaddlePaddle/PaddleOCR

Turn any PDF or image document into structured data for your AI. A powerful, lightweight OCR toolkit that bridges the gap between images/PDFs and LLMs. Supports 100+ languages.

95
/ 100
Verified

Built on the PaddlePaddle framework, PaddleOCR combines scene text detection, recognition, and layout analysis in an end-to-end pipeline that outputs structured formats (JSON, Markdown) optimized for LLM consumption. The toolkit supports heterogeneous hardware acceleration (CPU, GPU, NPU, Kunlun) and includes specialized models for handwriting and document understanding through its newer PaddleOCR-VL variant. It integrates with AI agents via MCP server protocol and has become a foundational dependency in document processing ecosystems like MinerU and RAGFlow.

72,167 stars and 1,622,419 monthly downloads. Used by 10 other packages. Actively maintained with 21 commits in the last 30 days. Available on PyPI.

Maintenance 23 / 25
Adoption 25 / 25
Maturity 25 / 25
Community 22 / 25

How are scores calculated?

Stars

72,167

Forks

9,954

Language

Python

License

Apache-2.0

Last pushed

Mar 12, 2026

Monthly downloads

1,622,419

Commits (30d)

21

Dependencies

4

Reverse dependents

10

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/rag/PaddlePaddle/PaddleOCR"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.