Marker

Apache-2.0 code, 38.7k GitHub stars, converts PDFs/DOCX/PPTX/XLSX to markdown/JSON

What Marker does

  • Apache-2.0 code, 38.7k GitHub stars, converts PDFs/DOCX/PPTX/XLSX to markdown/JSON
  • 76.0% accuracy on the olmocr-bench benchmark
  • Runs on GPU, CPU, or Apple Silicon; optional LLM-boosted accuracy mode

Marker — straight answers

What is Marker?

Marker is listed under AI OCR & Document Extraction, in the Research & Learning category on Flocci AI Tools. Apache-2.0 code, 38.7k GitHub stars, converts PDFs/DOCX/PPTX/XLSX to markdown/JSON. It is free, with no paid plan attached, and it lives at github.com.

Is Marker free?

Marker is listed as fully free — there is no paid tier attached to it in the catalog. That makes it one of the 396 entries on Flocci AI Tools with no upgrade path built in.

Can I use Marker in a browser, or is there an app?

Marker needs to be installed or self-hosted — it doesn't run in a browser tab. It is available on self-hosting, the command line and an API.

What can Marker do?

Marker does 3 things the catalog singles out: Apache-2.0 code, 38.7k github stars, converts pdfs/docx/pptx/xlsx to markdown/json; 76.0% accuracy on the olmocr-bench benchmark; runs on gpu, cpu, or apple silicon; optional llm-boosted accuracy mode.

What is the best free alternative to Marker?

dots.ocr (dots.mocr) is the closest free alternative: it sits in the same AI OCR & Document Extraction sub-category and is free. LangExtract (Google), MinerU and olmOCR also start free. The full list is on the alternatives page.

See the full list →

Marker alternatives

Compare all alternatives →

AskYourPDF

AI OCR & Document Extraction
freemium
  • Chat with any PDF plus auto-fill PDF forms
  • Embeddable document chatbot for websites
  • 5M+ users including MIT/Harvard/Oxford/Stanford

PDF.ai

AI OCR & Document Extraction
freemium
  • Chat-with-PDF product plus a separate parse/extract/split PDF API
  • Popular ChatPDF-category alternative with API access for developers

Humata

AI OCR & Document Extraction
freemium
  • Free tier: 60 pages, 10 answers, no card
  • Cited answers across many uploaded files at once
  • Paid tiers priced per extra page rather than per seat

Mistral OCR 4

AI OCR & Document Extraction
freemium
  • Self-hostable single-container OCR model for regulated industries
  • 170 languages, bounding boxes and per-field confidence scores
  • Low per-thousand-page API pricing; free to try inside Le Chat

LlamaParse

AI OCR & Document Extraction
freemium
  • Parses 90+ formats incl. complex tables/charts/handwriting into clean markdown
  • 300,000+ users, 1B+ documents processed
  • Free trial credits via LlamaCloud, then usage-based

Reducto

AI OCR & Document Extraction
freemium
  • 15,000 free starter credits, no card required
  • Citation-ready structured JSON across 30+ file types
  • Edits extracted data back into original PDF/DOCX layout

Head-to-head comparisons