Marker
Fast, accurate PDF-to-Markdown conversion.
About
Marker converts PDFs to Markdown (and JSON/HTML) quickly and accurately — the lightweight alternative for the most common document-ingestion job. 40.4k stars, Apache-2.0, 116k PyPI downloads/month, actively maintained. It trades Docling's breadth (many formats, deep layout analysis) for speed and simplicity on the PDF→Markdown path that feeds most RAG pipelines. Choose Marker when PDFs are the job; choose Docling when documents are diverse.
What it can do
- **Fast PDF→Markdown**: optimized for the common case
- **Accurate**: strong on text, tables, and structure
- **JSON/HTML output**: structured alternatives
- **Batch processing**: convert directories
- **Apache-2.0**: permissive
Setup
```bash
pip install marker-pdf
marker_single report.pdf --output_dir ./out
```
Docs in the repo.
Prerequisites
Python 3.9+; GPU speeds things up significantly.
Use cases
Works with
Trust score — 8/10
Scored 8/10 (verified tier, metrics-v1). Evidence: 40,409 GitHub stars (log-scaled vs catalog peers); blended with ~116,251 monthly package downloads; last push 11d ago; 2 releases/12mo; 477 open issues. Manual review: fast PDF→Markdown (40.4k stars, 116k PyPI/mo); Docling has broader formats and momentum. License: Apache-2.0.
★ 40,409 GitHub stars
How the trust score is built
Full methodology →- Adoption · stars/downloads, log-scaled vs peers
- 8.3/10
- Maintenance · recency, releases, issue load
- 7.2/10
- Openness · license + docs
- 10.0/10
- Manual review · sanity read + red-flag check
- 8.0/10
- · 40,409 GitHub stars (log-scaled vs catalog peers)
- · blended with ~116,251 monthly package downloads
- · last push 11d ago; 2 releases/12mo; 477 open issues
- · manual review: fast PDF→Markdown (40.4k stars, 116k PyPI/mo); Docling has broader formats and momentum
Sources
Every fact above was checked against these sources. Where a claim comes from the provider and couldn't be independently verified, the text says so.