datalab-to /marker

Convert PDF to markdown + JSON quickly with high accuracy

★ 40,926⑂ 2,954👁 138⑃ 412● 714▤ 74
PythonApache-2.0Created 2023-10-30www.datalab.to
More from this author
Download: Marker v2.2.0 — 4 Conversion Modes, In-Process Indra ONNX Engine, Interactive Web Viewer & 88.9% on OmniParseBenchMarker-2.2.0 ↓2.2.02026-11-12
+16
Daily
+102
Weekly
40,926
stars
2,954
forks
138
watchers
412
prs
714
issues
74
releases

README

Datalab

  State of the Art models for Document Intelligence


  
  
  


  
  
  




 Marker

Marker converts documents to markdown, JSON, chunks, and HTML quickly and accurately.

- Converts PDF, image, PPTX, DOCX, XLSX, HTML, EPUB files in all languages
- Formats tables, forms, equations, inline math, links, references, and code blocks
- Extracts and saves images
- Removes headers/footers/other artifacts
- Extensible with your own formatting and logic
- Optionally boost accuracy with LLMs (and your own prompt)
- Works on GPU, CPU, or MPS

 Try Datalab's Managed Platform

Our managed platform uses Chandra, our latest open source model, and other models — higher accuracy than Marker, with opt-in zero data retention, SOC 2 Type 2, and custom BAAs.

If you have high volume workloads, we offer a batch processing service that has processed 1B+ pages in a day at peak — we manage the infrastructure so your workloads finish on time.

The Datalab API turns documents into verifiable knowledge your agents act on. Every result traces back to its page and position, and one API can parse, extract, fill and edit documents.

Get started with about 5,000 free convert pages every month (work email) — sign up.

 Performance



Marker is built around a simple premise: you shouldn't have to choose between VLM-level accuracy and production throughput, or maintain separate tools for GPU and CPU.

On OmniParseBench (GitHub) — 16,288 pass/fail unit tests across 2,937 pages in 92 languages covering text & reading order (8,499 tests), table cell structure (5,386 tests), and block layout (2,403 tests) — as well as olmOCR-bench, Marker defines the Pareto frontier across every hardware tier:

- Faster and more accurate than open document pipelines (MinerU, docling, PP-StructureV3):
  - Marker balanced (80.5 OmniParseBench, 85.6 on the olmOCR-bench suites, 16.9 pg/s) beats docling (67.5 at 2.1 pg/s) by +13.0 points overall (+34.8 points on text & math) at 8× the speed, and runs 31× faster than MinerU 2.5-Pro (0.54 pg/s).
  - Marker vllm (85.9 OmniParseBench, 83.7 olmOCR-bench, 4.5 pg/s) delivers higher text accuracy (83.8 vs 77.4) and higher layout accuracy (95.7 vs 91.7) than MinerU 2.5-Pro at 8.3× the throughput (4.5 vs 0.54 pg/s).
- Outperforms open VLMs 2–12× its size (GLM-OCR, LightOnOCR-3-4B, PaddleOCR-VL-1.6, DeepSeek-OCR-2, olmOCR-2):
  - Full-page VLMs burn GPU compute on every single page — even clean born-digital PDFs. Marker's hybrid balanced mode keeps clean prose on CPU (pdftext + in-process Indra ONNX layout) and only calls the GPU VLM on equations, complex tables, CJK/non-Latin scripts, and scans — outscoring PaddleOCR-VL-1.6 (78.7) and DeepSeek-OCR-2 (75.2, 3.4B) at 14–21× higher throughput.
  - When you want full VLM OCR on every page, Marker vllm (powered by the compact 0.69B Surya 2.5 VLM) scores 85.9, beating 1B–8B open VLMs (LightOnOCR-3-4B 82.9, GLM-OCR 81.5, PaddleOCR-VL-1.6 78.7, DeepSeek-OCR-2 75.2, HunyuanOCR-1.5 64.6, olmOCR-2 42.6).
- Best-in-cl

Top issues

Latest release

Marker v2.2.0 — 4 Conversion Modes, In-Process Indra ONNX Engine, Interactive Web Viewer & 88.9% on OmniParseBench v2.2.0 2026-11-12
Marker v2.2.0 — 4 Conversion Modes, In-Process Indra ONNX Engine, Interactive Web Viewer & 88.9% on OmniParseBench

Marker 2.2.0 is our biggest architecture and usability upgrade since Marker 2.0:

- Up to 20× faster CPU conversion (7–33 pages/s) with zero external servers via the new in-process Indra ONNX layout, reading order, and OCR engine (fast and no_ocr modes).
- 88.9% on OmniParseBench in balanced mode — routing clean digital pages through pdftext + Indra ONNX and routing tables, math, CJK/non-Latin scripts, and scanned pages to Surya 2.5 VLM (86.7% in pure vllm mode), outperforming Do

Snapshot: 2026-11-16T09:23:14Z · Data license Apache-2.0 · JSON