Updated

Docling alternatives

For document parsing, the alternatives to Docling that makers' products we track chose most are Reducto (11), Unstructured (1) and Carbon (1). 9 of the 14 are open source or self-hostable (Unstructured, MarkItDown, Marker and 6 more) and 1 has a free tier.

9 open source or self-hostable1 with a free tier1 ready for AI agents

Not a ranking: each option is described by the situation it fits, with its trade-off.

Docling alternatives compared

How many makers' products we track use each for document parsing.

OptionKindMakers' productsFree tierOpen source
DoclingOpen-source library0—Yes
ReductoHosted API11—No
UnstructuredOpen-source library + hosted API1—Yes
MarkItDownOpen-source library0—Yes
MarkerOpen-source library + hosted API0—Yes
LlamaParseHosted API0—No
Mistral AIHosted API0YesNo
PyMuPDFPlain text extraction0—Yes
pypdfPlain text extraction0—Yes
pdf-parsePlain text extraction0—Yes
CarbonHosted API1—No
VectaraHosted API0NoSelf-hostable
ChunkrOpen-source library + hosted API0—Yes
R2ROpen-source library + hosted API0—Yes
WhyHow.AIOpen-source library + hosted API0—No

Hosted API or library you run. Hosted parsers handle messy scans, complex tables and odd layouts per page, with your documents going to their servers. Open-source libraries run locally for free and keep files private, but you provide the compute and accept lower accuracy on hard documents.

Docling (Open-source library): Parsing PDFs and office files locally, with layout and table structure kept, at no cost. Trade-off: Runs on your CPU or GPU, and heavy PDFs are slow without a GPU.

Reducto

Hosted API

Best forHard documents — scanned forms, dense tables, financial and legal files — where extraction accuracy matters most.

Trade-offPaid per page after starter credits, closed source, and documents leave your servers.

Used by 11 makers' products for this

Unstructured

Open-source library + hosted API
Open source

Best forMany file types through one pipeline, starting with the open-source library and moving to the hosted API later.

Trade-offThe open-source library is more basic than the hosted service, which bills per page.

Used by 1 maker's product for this

MarkItDown

Open-source library
Open source

Best forConverting PDFs, Word, Excel, PowerPoint, HTML and more to Markdown with one small Python package.

Trade-offLight on layout analysis, so complex tables and scans come out worse than with model-based parsers.

No maker product tracked for this yet

Marker

Open-source library + hosted API
Open source

Best forAccurate PDF-to-Markdown with layout, tables, equations and OCR from local models, or the same through Datalab's hosted API.

Trade-offThe model weights are free only for research, personal use and startups under $5M in funding or revenue; slow without a GPU.

No maker product tracked for this yet

LlamaParse

Hosted API

Best forHosted parsing that uses layout models and LLMs on hard pages, with modes you trade off between cost and accuracy.

Trade-offCredit-based pricing where the accurate modes cost several times the basic one, and documents go to LlamaIndex's cloud.

No maker product tracked for this yet

Mistral AI

Hosted API
Free tierFrom Pay per tokenllms.txtIn Replit

Best forOCR of scans and PDFs into Markdown with tables and images, from the same API as Mistral's text models.

Trade-offIt returns the page contents as OCR output, and pulling structured fields out is a second step with a model.

No maker product tracked for this yet

PyMuPDF

Plain text extraction
Open source

Best forVery fast PDF text, image and table extraction in Python, with PyMuPDF4LLM for Markdown output.

Trade-offAGPL-licensed, so closed-source products need a commercial license from Artifex.

No maker product tracked for this yet

pypdf

Plain text extraction
Open source

Best forPulling text out of born-digital PDFs in pure Python with no native dependencies.

Trade-offNo OCR and little layout awareness, so tables and multi-column pages come out jumbled.

No maker product tracked for this yet

pdf-parse

Plain text extraction
Open source

Best forExtracting text from PDFs in a Node.js app with one dependency.

Trade-offNo OCR, and layout-heavy pages lose their structure.

No maker product tracked for this yet

Carbon

Hosted API

Best forSyncing users' Google Drive, Notion and other sources into your vector database through one API.

Trade-offcarbon.ai no longer resolves and its SDK stopped in December 2024.

Used by 1 maker's product for this

Vectara

Hosted API
Self-hostableFrom $100K/year

Best forLarge companies that want ingestion, retrieval and cited answers as one managed platform, in SaaS, their VPC or on-premises.

Trade-offAfter a 30-day trial, SaaS starts at $100K a year (VPC $250K, on-prem $500K), far beyond a small team.

No maker product tracked for this yet

Chunkr

Open-source library + hosted API
Open source

Best forTurning PDFs, spreadsheets and images into LLM-ready Markdown, HTML or JSON with layout detection and bounding boxes.

Trade-offThe hosted API shuts down on 2026-11-15 as Chunkr joins Extend; self-hosting means AGPL terms.

No maker product tracked for this yet

R2R

Open-source library + hosted API
Open source

Best forA self-hosted agentic RAG server with ingestion, hybrid search and knowledge graphs behind a REST API.

Trade-offNo release since mid-2025 and its hosted site and docs are offline; apparently abandoned.

No maker product tracked for this yet

WhyHow.AI

Open-source library + hosted API

Best forTurning unstructured documents into knowledge graphs for more explainable retrieval.

Trade-offPivoted to legal-intelligence agents in 2025, then went offline; its GitHub organisation is gone.

No maker product tracked for this yet

Questions

What is the best alternative to Docling?

It depends on what you need it for; we don't rank them. The one makers' products we track chose most for document parsing is Reducto (11 makers' products). You only need the text out of born-digital PDFs, fast and free: pypdf, pdf-parse and PyMuPDF. Your inputs mix PDFs, emails, HTML and office files and you may move to a hosted API later: Unstructured. Scans, dense tables or forms where a parsing error means a wrong answer: Reducto, LlamaParse and Mistral AI.

Is there a free alternative to Docling?

Yes. For document parsing, Mistral AI has a free tier a small product can start on.

Is there an open-source alternative to Docling?

Yes. For document parsing: Unstructured, MarkItDown, Marker, PyMuPDF, pypdf, pdf-parse and 3 more.