← back to the directory
Library / SDK Retrieval & Memory

Marker

1Fast 2accurate 3PDF 4to 5markdown 6conversion

the six Ws · specification

W1 Who

Developers needing fast, accurate local conversion of PDFs and Office files into Markdown.

W2 What

Marker is an open-source tool that converts PDF, DOCX, PPTX, and other documents into Markdown, JSON, or HTML using its own layout model and Surya OCR.

W3 Where

Self-hosted as a Python package (pip install marker-pdf) run on local or server CPUs/GPUs.

W4 When

Rewritten as Marker 2 for speed and CPU support, with active releases through 2026.

W5 Why

It exists to give developers a fast, self-hostable alternative to paid document-parsing APIs.

W6 With

Built on Datalab's Surya OCR and its own pdftext extraction engine.

W7 Watch

Open source, with a separate commercial license track by Datalab; self-hosted; no data leaves local infrastructure unless using Datalab's hosted API.

PDF to markdownlayout modelSurya OCRfast conversion

for agents & scripts

Reading this as a machine? Query it directly.

Search is open JSON - no key. Report telemetry after using a tool and it feeds that tool’s Proof Score. Or speak MCP to /mcp and discover tools mid-loop.