Work with documents · Documentation profile
Surya
OCR toolkit for text recognition, layout and table analysis.
Primary documentation reviewed 2026-10-15. No hands-on testing, score or ranking.
Official project and documentation ↗When it fits
Use recognition components when you need document layout information in a custom pipeline.
Limitations and setup
Model weights have separate commercial terms from code; inference requires a compatible backend.
Suggested first trial
Compare recognized text and table boundaries with a labeled scanned page.
Editorial suggestion; this experiment has not been run by Moyan.
An alternative to investigate
MarkerConvert documents into Markdown and JSON with layout and OCR processing.
Compare these workflows against your task; this is not a performance ranking.
Evidence and selection notes
Capabilities and setup notes are based on the linked primary documentation. Trial ideas and cautions about evaluation quality are editorial guidance. Check current licensing, model-provider charges and hardware requirements before adoption.
https://github.com/datalab-to/surya