OLMo

Health v2 · maintenance only

Adoption not part of the score

6692 stars · 799 forks observed · 2026-10-07

What it is AI-extracted, prompt v1, taxonomy v1, 2026-10-19, confidence 0.97

olmOCR is a Python library that converts PDFs and image-based documents into clean, readable plain text or Markdown using a visual language model. It supports equations, tables, handwriting, and complex layouts while removing headers and footers.

Use cases

• extract clean text from research PDFs for LLM training
• convert scanned documents and images into markdown for data pipelines
• process multi‑column academic papers into a linear reading order
• remove headers/footers and preserve tables and equations in OCR output
• generate high‑quality text from PDFs for downstream NLP tasks

When to choose

• you need high‑quality OCR that preserves complex formatting like tables and equations
• your workflow requires converting large volumes of PDFs to plain text or markdown efficiently
• you have GPU resources and want to leverage a visual language model for better accuracy
• you want an open‑source, Apache‑2.0 licensed solution with active development

When to avoid

• you lack GPU access or need a CPU‑only solution
• you require real‑time OCR on low‑power devices
• your documents are simple text images where a basic OCR engine would suffice
• you need a fully managed SaaS OCR service with built‑in scaling

Facets

library · maturity active

ocr image-processing parser pdf nlp computer-vision data-science python windows open-source llm-dataset markdown-output gpu-accelerated vllm-inference natural-language-processing linux macos docker gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="allenai/OLMo")