An Unstructured alternative that runs in your own code.
PDFluent is an Unstructured alternative that runs inside your own process on your own machine, with no cloud service. We compare where each runs, benchmark scores and list prices, and we read competitor facts between 2 and 16 November 2026.
Competitor facts read between 2 and 16 November 2026.
PDFluent and Unstructured, row by row.
The benchmark's README lists unstructured hi_res at 0.841 and unstructured at 0.686 overall, against 0.9282 for PDFluent 1.4.0 quality auto. PDFluent is ahead on all 4 scores: overall, reading order, table structure and heading hierarchy. This is one run and one benchmark. The competitor figures come from that README, and the rows are the open-source library as the leaderboard lists it.*
| Dimension | PDFluent | Unstructured |
|---|---|---|
| Where it runs | Inside your own process, on your own machine. No cloud service, licence server or activation. | Hosted SaaS, or a dedicated instance, in-VPC or bare metal on the Business plan. The open-source library runs locally. |
| Licence | AGPL-3.0, or a commercial licence | Open-source library: Apache-2.0. Hosted API: free tier and paid plans. |
| Price model | SDK: no per-page fee. Hosted API: coming soon, prepaid credit packs. | 10,000 free pages to start, then $0.015 per page |
| opendataloader-bench, overallPDFluent ahead | 0.9282 (PDFluent 1.4.0, quality auto) | 0.841 (unstructured [hi_res]), 0.686 (unstructured) |
| Reading orderPDFluent ahead | 0.9463 | 0.904 (hi_res), 0.882 |
| Table structure (TEDS)PDFluent ahead | 0.9140 | 0.588 (hi_res), 0.000 |
| Heading hierarchyPDFluent ahead | 0.8767 | 0.749 (hi_res), 0.388 |
| Connectors and file typesOther tool ahead | In Python: CSV, TSV, JSON, XML, RTF, HTML, Markdown, plain text and e-mail next to PDF, plus LangChain and LlamaIndex loaders | 40+ connectors (20+ sources, 20+ destinations) and 50+ document and image file types |
| Hosted API todayOther tool ahead | Not yet. Coming soon. | Yes, SaaS with a free tier |
* Overall score 0.9282 (SDK 1.4.0, quality="auto", keep_furniture=True, layout, table and formula models installed) on all 200 documents, with the benchmark’s own harness and evaluator, measured 11 November 2026 on the final macOS arm64 release wheel. Competitor figures are taken from the benchmark’s public README as read on 3 November 2026 (highest overall there: 0.907, opendataloader [hybrid]), not re-run by us. The leaderboard engines ran on their own setups while PDFluent was measured on macOS arm64, and table structure depends on the platform. PDFluent is not on that leaderboard. Other engines have submissions open that are not yet on the leaderboard. One run, no confidence interval. The PDFluent row keeps page headers and footers because the benchmark’s ground truth counts them as content. PyMuPDF and MuPDF are trademarks of Artifex Software, Inc. PDFluent is an independent product of Innovation Trigger B.V. and is not affiliated with or endorsed by Artifex.
Every engine, one scale.
- PDFluent 1.4.00.9282
- opendataloader [hybrid]0.907
- nutrient0.885
- docling0.882
- marker0.861
- unstructured [hi_res]0.841
- edgeparse0.837
- opendataloader0.831
- mineru0.831
- pymupdf4llm0.732
- unstructured0.686
- markitdown0.589
- liteparse0.576
PDFluent 1.4.0 with quality auto, page headers and footers kept. Other engines as published in the benchmark README on 2 November 2026. Footnote
The chart shows the 12 engines on the public opendataloader-bench leaderboard next to PDFluent 1.4.0 at quality auto. On the four scores of that benchmark PDFluent is ahead of Unstructured on 4 of 4. The table under the chart gives every engine on every score.
PDFluent and the 12 leaderboard engines on all four scores
| Engine | Overall | Reading order | Table structure | Headings |
|---|---|---|---|---|
| PDFluent 1.4.0 | 0.9282 | 0.9463 | 0.9140 | 0.8767 |
| opendataloader [hybrid] | 0.907 | 0.934 | 0.928 | 0.821 |
| nutrient | 0.885 | 0.925 | 0.708 | 0.819 |
| docling | 0.882 | 0.898 | 0.887 | 0.824 |
| marker | 0.861 | 0.890 | 0.808 | 0.796 |
| unstructured [hi_res] | 0.841 | 0.904 | 0.588 | 0.749 |
| edgeparse | 0.837 | 0.894 | 0.717 | 0.706 |
| opendataloader | 0.831 | 0.902 | 0.489 | 0.739 |
| mineru | 0.831 | 0.857 | 0.873 | 0.743 |
| pymupdf4llm | 0.732 | 0.885 | 0.401 | 0.412 |
| unstructured | 0.686 | 0.882 | 0.000 | 0.388 |
| markitdown | 0.589 | 0.844 | 0.273 | 0.000 |
| liteparse | 0.576 | 0.866 | 0.000 | 0.000 |
Their call, next to ours.
1import os23import unstructured_client4from unstructured_client.models import operations, shared56client = unstructured_client.UnstructuredClient(7 api_key_auth=os.getenv("UNSTRUCTURED_API_KEY")8)910filename = "report.pdf"1112req = operations.PartitionRequest(13 partition_parameters=shared.PartitionParameters(14 files=shared.Files(15 content=open(filename, "rb"),16 file_name=filename,17 ),18 strategy=shared.Strategy.VLM,19 vlm_model="gpt-4o",20 vlm_model_provider="openai",21 ),22)2324res = client.general.partition(request=req)25elements = [element for element in res.elements]
1from pdfluent import Document2from pdfluent.chunks import iter_chunks34# pip install pdfluent5with Document("report.pdf") as doc:6 markdown = doc.extract_markdown()7 for chunk in iter_chunks(doc, mode="markdown"):8 print(chunk.text)
Unstructured returns a list of elements. PDFluent returns Markdown for the document, or chunks with their text, so the step that groups elements into chunks can go. The Unstructured code is the example from its own documentation, read on 16 November 2026.
What each costs, as of 16 November 2026.
We read the list prices below on 16 November 2026. Unstructured charges $0.015 per page on Pay-As-You-Go after the first 10,000 free pages. PDFluent is a licence for software you run yourself at EUR 999 per organisation per year, excluding VAT, as of 7 September 2026.
| Product | Plan | Price |
|---|---|---|
| PDFluent SDK | Free under the AGPL-3.0, or a commercial licence | EUR 0, or EUR 999 per organisation per year, excluding VAT (as of 7 September 2026)No per-page fee. You run it on your own hardware. |
| PDFluent hosted API | Coming soon, not live yet | Planned: prepaid packs from USD 25 for 25,000 credits. 1 credit per page at fast, 3 at autoUSD 1.00 and USD 3.00 per 1,000 pages at the USD 25 pack. Planned terms as of 16 November 2026. |
| Unstructured | Free | 10,000 free pages to start, no card required |
| Unstructured | Pay-As-You-Go | $0.015 per page after the first 10,000 pages, which is $15.00 per 1,000 pagesOne per-page rate on the page; no separate rate per strategy. |
| Unstructured | Business | Custom pricing |
* List prices read on 16 November 2026 from each vendor’s own pricing page, and the arithmetic on them. The services name their tiers and strategies differently and have not been measured against each other, so a lower price is not a like for like saving. The PDFluent hosted API is coming soon and is not live; its prices are planned terms, checked again at launch. The SDK runs on your own hardware, which has its own cost. The SDK price is as of 7 September 2026.
Who each one suits.
Pick PDFluent when
PDFluent fits teams that want the work to stay on their own machines. The SDK runs inside your own process, needs no cloud service, licence server or activation, and its default build has no HTTP client. It ships bindings for Rust, Python, Node.js, WebAssembly, Java and .NET, plus a command-line tool. You can use it free under the AGPL-3.0, or take a commercial licence if you do not want to publish your own source. Structured extraction returns Markdown, blocks, words, Docling JSON and chunks in Python.
Pick Unstructured when
Unstructured fits teams that want a hosted service today, with a Free plan and Pay-As-You-Go SaaS. The Business plan adds custom pricing, a dedicated instance, VPC or multi-tenant SaaS. Unstructured lists 40+ connectors, 20+ sources and 20+ destinations. It supports 50+ document and image file types. Its strategies are Auto, VLM, High-Res and Fast. Bare metal is another Business plan deployment option.
Three questions.
What does each cost?+
Is there a hosted PDFluent API?+
How hard is it to switch?+
Try it on a
document you know.
Install it, open one file you can judge, and read what comes back.
Other comparisons: LlamaParse alternative · Docling alternative