TOOLORA / Local reading edition
Marker
Convert PDFs and documents to Markdown, JSON or chunks, with optional model-assisted refinement.
Overview
Marker is Datalab’s document converter for text, tables, equations and images from PDFs and other formats. Fast and balanced paths combine text layers, layout models, OCR and vision models; optional LLM calls refine difficult results. Datalab’s managed platform uses Chandra, a different model, so its performance or data-handling claims cannot automatically be attributed to Marker.
- Company country / region
- United States · Published company address
Usage guide & costs
- Before you start
Code is Apache 2.0; weights use modified OpenRAIL-M. Weight terms restrict non-personal/non-research use above US$5M prior-year revenue or total funding, and separately restrict competing products/services. Check the full license; optional external LLMs have separate data/cost terms.
Official details ↗ · Official details ↗ · Official details ↗
Official sources and references
- Official websitehttps://github.com/datalab-to/marker
- Documentationhttps://github.com/datalab-to/marker/blob/master/README.md
Descriptions summarize public information; consult official websites for features, pricing and regional availability.