LlamaIndex releases LiteParse v2.0, a Rust rewrite claiming up to 100x faster document parsing

VIEWS398.6KBOOKMARKS3.5KLIKES3.4KRETWEETS298REPLIES80

We've created the world's fastest PDF parser ⚡️

And it's more accurate than any other open-source, model-free PDF parser out there (pymupdf, pypdf, markitdown, pdftotext, opendataloader, pymupdf4llm)

Introducing LiteParse v2 - we rewrote the entire library into Rust and adapted it as native packages for Python and Node.

It supports 50+ different document types, can be triggered directly or installable directly within your favorite AI agent.

Blog: https://www.llamaindex.ai/blog/liteparse-v2-0-runs-everywhere?utm_medium=socials&utm_source=twitter&utm_campaign=2026-may-

Repo: https://github.com/run-llama/liteparse

LlamaIndex 🦙@llama_index

LiteParse v2.0 is out now, and it is blazing fast + runs everywhere!

We rewrote everything from scratch in Rust, and now: - up to 100x faster parsing - install natively in Rust, JS/TS, and Python - a custom WASM package enables browser and edge runtime usage

pip install liteparse npm i ＠llamaindex/liteparse npm i ＠llamaindex/liteparse-wasm cargo install liteparse

Blog: https://www.llamaindex.ai/blog/liteparse-v2-0-runs-everywhere?utm_medium=socials&utm_source=twitter&utm_campaign=2026-may- Repo: https://github.com/run-llama/liteparse

34d398.6K3.4K3.5K

Jerry Liu@jerryjliu0

Parse PDFs at lightspeed (this video is at 1x)

Absolute cinema

Jerry Liu@jerryjliu0

We've created the world's fastest PDF parser ⚡️

And it's more accurate than any other open-source, model-free PDF parser out there (pymupdf, pypdf, markitdown, pdftotext, opendataloader, pymupdf4llm)

Introducing LiteParse v2 - we rewrote the entire library into Rust and adapted it as native packages for Python and Node.

It supports 50+ different document types, can be triggered directly or installable directly within your favorite AI agent.

Blog: https://www.llamaindex.ai/blog/liteparse-v2-0-runs-everywhere?utm_medium=socials&utm_source=twitter&utm_campaign=2026-may-

Repo: https://github.com/run-llama/liteparse

32d123.9K833867

Jerry Liu@jerryjliu0

The secret to LiteParse lies in the grid projection algorithm. We project a complex page layout with text and tables into well-structured text, that humans can read and agents can understanding.

This contains of a few core steps (no LLMs!): 1. Grouping text fragments to lines 2. Identify left,center,right anchors 3. Snap each text item to an anchor 4. Handle flowing paragraphs separately 5. Render each text item in a carefully tuned order so that each piece of text aligns to a grid column 6. Post-processing

For more details check out this great blog post we wrote a month ago!

https://www.llamaindex.ai/blog/how-liteparse-turns-pdfs-into-text-a-deep-dive-into-the-grid-projection-algorithm

Jerry Liu@jerryjliu0

We've created the world's fastest PDF parser ⚡️

And it's more accurate than any other open-source, model-free PDF parser out there (pymupdf, pypdf, markitdown, pdftotext, opendataloader, pymupdf4llm)

Introducing LiteParse v2 - we rewrote the entire library into Rust and adapted it as native packages for Python and Node.

It supports 50+ different document types, can be triggered directly or installable directly within your favorite AI agent.

Blog: https://www.llamaindex.ai/blog/liteparse-v2-0-runs-everywhere?utm_medium=socials&utm_source=twitter&utm_campaign=2026-may-

Repo: https://github.com/run-llama/liteparse

31d22.7K181132

Jerry Liu@jerryjliu0

Beyond being fast, LiteParse is designed to provide highly accurate, semantically coherent text for LLM use.

We benchmarked every open-source, model-free PDF parser on LLM QA tasks - from PyPDF to PyMuPDF to Markitdown.

✅ We ~roughly tied for #1 in accuracy (along with pdftotext, which is decently accurate but a bit slower)

✅ PyMuPDF is the closest to us in term of latency, but we found it struggles in projecting complex text layouts (multi-columns, tables) in formats that LLMs can understand

Besides being accurate and #1 in speed, LiteParse is also a general-purpose parser taht supports dozens of other file formats (incl .docx, .pptx, .xlsx), and also supports convenience tools for both OCR and screenshotting.

Come check it out!

LiteParse: https://github.com/run-llama/liteparse

Jerry Liu@jerryjliu0

We've created the world's fastest PDF parser ⚡️

And it's more accurate than any other open-source, model-free PDF parser out there (pymupdf, pypdf, markitdown, pdftotext, opendataloader, pymupdf4llm)

Introducing LiteParse v2 - we rewrote the entire library into Rust and adapted it as native packages for Python and Node.

It supports 50+ different document types, can be triggered directly or installable directly within your favorite AI agent.

Blog: https://www.llamaindex.ai/blog/liteparse-v2-0-runs-everywhere?utm_medium=socials&utm_source=twitter&utm_campaign=2026-may-

Repo: https://github.com/run-llama/liteparse

33d21.8K199105

Dmitry Lyalin@LyalinDotCom

This is pretty cool!

Jerry Liu@jerryjliu0

We've created the world's fastest PDF parser ⚡️

And it's more accurate than any other open-source, model-free PDF parser out there (pymupdf, pypdf, markitdown, pdftotext, opendataloader, pymupdf4llm)

Introducing LiteParse v2 - we rewrote the entire library into Rust and adapted it as native packages for Python and Node.

It supports 50+ different document types, can be triggered directly or installable directly within your favorite AI agent.

Blog: https://www.llamaindex.ai/blog/liteparse-v2-0-runs-everywhere?utm_medium=socials&utm_source=twitter&utm_campaign=2026-may-

Repo: https://github.com/run-llama/liteparse

33d8.8K4028

simon@disiok

we are a car

Jerry Liu@jerryjliu0

We've created the world's fastest PDF parser ⚡️

And it's more accurate than any other open-source, model-free PDF parser out there (pymupdf, pypdf, markitdown, pdftotext, opendataloader, pymupdf4llm)

Introducing LiteParse v2 - we rewrote the entire library into Rust and adapted it as native packages for Python and Node.

It supports 50+ different document types, can be triggered directly or installable directly within your favorite AI agent.

Blog: https://www.llamaindex.ai/blog/liteparse-v2-0-runs-everywhere?utm_medium=socials&utm_source=twitter&utm_campaign=2026-may-

Repo: https://github.com/run-llama/liteparse

33d7.7K2016

Jerry Liu@jerryjliu0

Also, it's completely open-source and free!

Jerry Liu@jerryjliu0

We've created the world's fastest PDF parser ⚡️

And it's more accurate than any other open-source, model-free PDF parser out there (pymupdf, pypdf, markitdown, pdftotext, opendataloader, pymupdf4llm)

Introducing LiteParse v2 - we rewrote the entire library into Rust and adapted it as native packages for Python and Node.

It supports 50+ different document types, can be triggered directly or installable directly within your favorite AI agent.

Blog: https://www.llamaindex.ai/blog/liteparse-v2-0-runs-everywhere?utm_medium=socials&utm_source=twitter&utm_campaign=2026-may-

Repo: https://github.com/run-llama/liteparse

34d8.1K845

Clelia Bertelli (🦙/acc)@itsclelia

LiteParse 🤝 Rust🦀 We refactored the LiteParse library and CLI porting it to Rust, and here's what that means for you: ⚡ Parsing up to 100X faster, with sub-second processing for documents as large as 450+ pages and 100MB in size 🈳 Has native multi-language support, with bindings for python and typescript (and the rust library and CLI) 🌐 Can be truly deployed on edge devices and in the browser, as we cross-compiled it to WASM. We also have a fun demo for it 👉 http://run-llama.github.io/liteparse

Kudos to @LoganMarkewich for driving this!🎉

📚 Check out the blog: https://www.llamaindex.ai/blog/liteparse-v2-0-runs-everywhere ⭐ Star the repo (we crossed 5k!): https://github.com/run-llama/liteparse

33d1.9K1911

Jonathan Rhyne@jdrhyne

@jerryjliu0 Would really re-run this on a commonly used open-source benchmark like opendataloader's that has all these parsers on there - https://github.com/opendataloader-project/opendataloader-pdf

I ran it and the speed increase is impressive (nice job on the Rust rewrite!) but the accuracy claim seems a bit off.

33d1.5K169

Jerry Liu@jerryjliu0

@aaryan_kakad @llama_index yes! https://github.com/run-llama/liteparse

31d507811

Jonathan Rhyne@jdrhyne

@davielam_ @jerryjliu0 Wonder how ours compares - https://github.com/PSPDFKit/pdf-to-markdown

33d81287

kogu@koguGameDev

https://github.com/run-llama/liteparse これか。日本語PDFでどんなか試しとこう。

Jerry Liu@jerryjliu0

We've created the world's fastest PDF parser ⚡️

And it's more accurate than any other open-source, model-free PDF parser out there (pymupdf, pypdf, markitdown, pdftotext, opendataloader, pymupdf4llm)

Introducing LiteParse v2 - we rewrote the entire library into Rust and adapted it as native packages for Python and Node.

It supports 50+ different document types, can be triggered directly or installable directly within your favorite AI agent.

Blog: https://www.llamaindex.ai/blog/liteparse-v2-0-runs-everywhere?utm_medium=socials&utm_source=twitter&utm_campaign=2026-may-

Repo: https://github.com/run-llama/liteparse

33d3.9K116

Jonathan Rhyne@jdrhyne

@davielam_ @jerryjliu0

33d78446

Davie@davielam_

✅ I reran the 23-PDF benchmark:

- Corpus: 23 local PDFs / 82 pages - invoices/labels/SDS/barcode PDFs/Hermes docs/research/sample scanned PDFs - Result: PSPDFKit/Nutrient still fastest - 4.58s total / 0.20s per PDF - ~40× faster than PyMuPDF total runtime - But: PSPDFKit had empty/tiny output on 7/23 - especially OCR/image/outline-heavy label/manual PDFs - Failures: no crashes - PSPDFKit empty: 7/23 - PyMuPDF empty: 1/23 - Docling empty: 2/23 - Decision unchanged: keep PyMuPDF as current safe default; use PSPDFKit only as fast first-pass with fallback.

33d76764

Jason@snellchapo

@jerryjliu0 You get smoked by pymupdf on everything except 457 pages..

33d61562

Davie@davielam_

I ran a test between yours vs PyMuPDF: - Winner: current Hermes local PDF path = PyMuPDF - liteparse: installed only in temp venv, benchmarked, then removed - Why not keep liteparse: ~2.1× faster, but worse reading order on two-column PDFs; it merges left/right columns into same lines, which is worse for LLM/document reading. - Benchmark set: 4 real Hermes PDFs, 52 pages total - Quality: liteparse word coverage was high 99.4–100%, but layout/readability lost. - Activation needed: none — liteparse was not kept.

33d35221

Guillermo Rauch@rauchg

@jerryjliu0 Cool

Jerry Liu@jerryjliu0

We've created the world's fastest PDF parser ⚡️

And it's more accurate than any other open-source, model-free PDF parser out there (pymupdf, pypdf, markitdown, pdftotext, opendataloader, pymupdf4llm)

Introducing LiteParse v2 - we rewrote the entire library into Rust and adapted it as native packages for Python and Node.

It supports 50+ different document types, can be triggered directly or installable directly within your favorite AI agent.

Blog: https://www.llamaindex.ai/blog/liteparse-v2-0-runs-everywhere?utm_medium=socials&utm_source=twitter&utm_campaign=2026-may-

Repo: https://github.com/run-llama/liteparse

33d3.1K121

Jamie Lemon@jamie_lemon_x

@jerryjliu0 PyMuPDF4LLM does have a lightweight model built in by default - see: https://www.pdf4llm.com , however you can turn this off if required: https://pymupdf.readthedocs.io/en/latest/pymupdf4llm/api.html#use_layout Did your results disable the model okay to make it truly lightweight for your comparison table?

33d61212

Kapil Dave@kapildave_

@llama_index The Rust-rewrite-of-parsing-tool playbook is becoming a pattern. CPU-bound primitives → Rust rewrite → 50-100x. Most of the productivity gain isn't language speed, it's the absence of Python's GC tail latency at high concurrency. Worth shipping.

33d1.1K61

Nick Craske@nickcraske

⚡️

Jerry Liu@jerryjliu0

We've created the world's fastest PDF parser ⚡️

And it's more accurate than any other open-source, model-free PDF parser out there (pymupdf, pypdf, markitdown, pdftotext, opendataloader, pymupdf4llm)

Introducing LiteParse v2 - we rewrote the entire library into Rust and adapted it as native packages for Python and Node.

It supports 50+ different document types, can be triggered directly or installable directly within your favorite AI agent.

Blog: https://www.llamaindex.ai/blog/liteparse-v2-0-runs-everywhere?utm_medium=socials&utm_source=twitter&utm_campaign=2026-may-

Repo: https://github.com/run-llama/liteparse

33d3.1K51