Complete pricing guide for Marker. Compare all plans, analyze costs, and find the perfect tier for your needs.
Not sure if free is enough? See our Free vs Paid comparison →
Still deciding? Read our full verdict on whether Marker is worth it →
forever
GPL license; model weights free for research, personal use, and startups under $2M
per use
Pay per page processed
annual
Contact DataLab for pricing
Pricing sourced from Marker · Last verified March 2026
Marker benchmarks favorably against both. Compared to LlamaParse, Marker is faster and open-source. Compared to Docling, Marker focuses on markdown/JSON output quality while Docling provides richer structured output with bounding boxes. Marker's LLM-enhanced mode often produces the highest overall accuracy.
Not technically — it runs on CPU and Apple MPS — but practically, yes for any batch workload. CPU processing is 4-10x slower. GPU processing achieves roughly 25 pages/second on H100 hardware in batch mode. For a handful of documents, CPU works fine.
Yes, through integrated Surya OCR supporting 90+ languages. Scanned documents at 300+ DPI produce good results. Use the --force_ocr flag to ensure all content goes through OCR. Lower-quality scans will have reduced accuracy.
The --use_llm flag pairs Marker's layout detection with an LLM (Gemini Flash by default) for post-processing. It improves table formatting, handles inline math, merges split tables, and extracts form values. Use it when document accuracy matters more than processing speed — it adds cost and latency but produces measurably better output.
AI builders and operators use Marker to streamline their workflow.
Try Marker Now →IBM-backed open-source document parsing toolkit that converts PDFs, DOCX, PPTX, images, audio, and more into structured formats for RAG pipelines and AI agent workflows.
Compare Pricing →LlamaParse: Extract and analyze structured data from complex PDFs and documents using LLM-powered parsing.
Compare Pricing →Document ETL engine that converts messy PDFs, Word files, and images into AI-ready structured data with intelligent chunking.
Compare Pricing →Enterprise-grade text extraction and document processing framework that detects and extracts content from 1,000+ file formats. Free, containerized, and battle-tested across 18 years of production deployment.
Compare Pricing →AWS document intelligence service that extracts text, tables, forms, and handwriting from scanned documents using machine learning — with specialized APIs for invoices, IDs, and lending documents.
Compare Pricing →