olmOCR

What is olmOCR?

olmOCR is a toolkit from the Allen Institute for AI (Ai2) for converting PDFs, PNG and JPEG documents into clean Markdown in natural reading order. It is built on a 7B vision-language model and handles equations, tables, handwriting, multi-column layouts and figures, and removes headers and footers. It runs batch jobs from the command line on a local NVIDIA GPU via vLLM, in Docker, or against any OpenAI-compatible inference server.

Where it fits in a 10–200 person business

• Convert scanned contracts, invoices and reports into Markdown for search or a RAG knowledge base
• Batch-digitise an archive of PDFs with tables and equations
• Extract text from handwritten forms and multi-column documents

These use cases are our editorial suggestions, not claims made by the project.

Can a business use olmOCR? (license)

This is a plain-English summary of the license text, not legal advice. Third-party models you run with it have their own licenses.

Data & compliance: does olmOCR phone home?

Checked against the project's own docs and repository on Nov 3, 2026. "Not documented" means we could not find the answer in official sources; it does not mean the feature is missing or that no data is sent. Third-party services you connect (model APIs, vector databases, search engines) have their own data policies. Compare all apps.

What hardware does olmOCR need?

The README lists 30 GB of free disk space for local GPU inference; no CPU or RAM figures are published.

Source: github.com/allenai/olmocr#python-installation

How much does it cost to host olmOCR per month?

How to deploy olmOCR

Method: Docker (GPU image with model included)

docker pull alleninstituteforai/olmocr:latest-with-model
docker run --gpus all \
  -v /path/to/pdfs:/input \
  -v /path/to/output:/output \
  alleninstituteforai/olmocr:latest-with-model \
  -c "olmocr /output --markdown --pdfs /input/*.pdf"

From the official instructions: github.com/allenai/olmocr#using-docker. Always check the source for the current version before you run it.

Known gotchas when self-hosting olmOCR

• The image that includes the model is about 30 GB; a smaller base image without the model is available if you manage model downloads yourself. source
• The Python requirements are hard to install into an existing environment; the project asks for a clean Conda environment (Python 3.11). source
• Using an external provider with --server sends your document pages to that provider; keep inference on your own GPU or vLLM server for private documents. source

Alternatives in Document processing

• PaddleOCR OCR toolkit and document AI engine that turns scans, images and PDFs into LLM-ready text, Markdown or JSON. Compare olmOCR vs PaddleOCR
• MinerU Document parser that converts PDFs, scans and Office files into clean Markdown or JSON for LLMs and RAG. Compare MinerU vs olmOCR
• Paperless-ngx Document management system that OCRs scans into a searchable archive, auto-tags them and adds optional LLM suggestions and chat. Compare olmOCR vs Paperless-ngx
• OCRmyPDF Command-line tool and Docker image that adds a searchable OCR text layer to scanned PDFs, with a watched-folder service. Compare OCRmyPDF vs olmOCR
• Unstract No-code LLM platform that turns invoices, forms and other documents into structured JSON. Compare olmOCR vs Unstract
• Paperless-AI AI auto-tagging and document chat for Paperless-ngx (currently unmaintained). Compare olmOCR vs Paperless-AI
• paperless-gpt LLM add-on for paperless-ngx that does better OCR and writes titles, tags and correspondents. Compare olmOCR vs paperless-gpt
• Docling Serve Docling document conversion (PDF, Office, images to Markdown/JSON) as a self-hosted API with a demo UI. Compare Docling Serve vs olmOCR

Sources

• GitHub repository: github.com/allenai/olmocr
• Official documentation: github.com/allenai/olmocr#readme
• License text: github.com/allenai/olmocr/blob/main/LICENSE
• Hardware requirements, Gotcha 2: github.com/allenai/olmocr#python-installation
• Deploy instructions, Gotcha 1: github.com/allenai/olmocr#using-docker
• Gotcha 3: github.com/allenai/olmocr#using-an-inference-provider-or-external-server
• Hetzner Cloud pricing (cost estimate): www.hetzner.com/cloud/regular-performance
• GitHub stars, commits and releases: GitHub REST API, fetched Nov 3, 2026