Skip to content

About

Jupyter notebooks for running DeepSeek-OCR batch PDF processing on Runpod

Resources

Stars

12 stars

Watchers

0 watching

Forks

Repository files navigation

DeepSeek-OCR Companion Files

Notebooks and examples related to our DeepSeek-OCR videos: https://www.youtube.com/@Integral_Business_Intelligence

What's Here

  • DeepSeek-OCR Research Paper - The original paper from DeepSeek
  • 00_setup_environment.ipynb - Runpod environment setup notebook
  • 01_vllm-two-pass.ipynb - Two-pass OCR pipeline for batch PDF processing
  • deepseek_ocr_paper.md - The DeepSeek-OCR paper processed through its own model

Usage

  1. Run the setup notebook to prepare your environment
  2. Upload PDFs to /workspace/pdfs
  3. Run the processing notebook

The pipeline does two-pass OCR with figure extraction and description generation.

Notes

These are companion materials for our YouTube channel. The notebooks assume a Runpod-style environment but can be adapted to other setups.

About

Jupyter notebooks for running DeepSeek-OCR batch PDF processing on Runpod

Resources

Stars

12 stars

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages