TechPrufer
Get started
Replicate
Replicate logoReplicate · Feature trail

Document Parsing and OCR

1 post · Oct 2025

What this trail covers

Replicate now offers advanced document parsing and text extraction capabilities through the integration of Datalab Marker and OCR models. Marker processes various document formats (PDF, DOCX, PPTX, images) into markdown or JSON, handling tables, math, and code, and supporting structured field extraction via JSON Schema. OCR detects text in 90 languages from images and documents, providing reading order and table grids. Both models offer high performance, outperforming established tools like Tesseract, with short processing times. Marker processes a page in about 0.18 seconds and can hit 120 pages per second when batched. Structured extraction is a powerful feature of Marker, allowing specific fields to be extracted from documents like invoices. Performance evaluations show Marker outperforming other models on benchmarks like olmOCR-Bench.

All Replicate trails
  1. Extract text from documents and images with Datalab Marker and OCRThis post introduces the integration of Datalab Marker and OCR models on Replicate. It details the capabilities of Marker for parsing various document formats into markdown or JSON, including handling tables, math, and code, and supporting structured field extraction via JSON Schema. It also describes the OCR model's ability to detect text in 90 languages from images and documents, providing reading order and table grids. Performance benchmarks are presented, showing Marker outperforming established tools like Tesseract and GPT-4o on the olmOCR-Bench dataset. Pricing for both models is also detailed.