LiteParse V2.1 Launches Fastest LLM-Free Markdown Output

Original post

LiteParse v2.1 is here, and its bringing the fastest markdown output possible.

In this release, we are fulfilling our top request: markdown output. But in the spirit of "lite"-ness, we are doing this completely LLM-free and fast.

Not only is it fast, it also beats all other model-free competitors in 3 separate benchmark datasets.

Read more about it in our release blog: https://www.llamaindex.ai/blog/markdown-comes-to-liteparse?utm_medium=socials&utm_source=twitter&utm_campaign=2026-jun-

10:17 AM · Jun 18, 2026 · 19.8K Views

VIEWS18.8KBOOKMARKS170LIKES132RETWEETS24REPLIES10

Jerry Liu@jerryjliu0

We built the fastest PDF -> markdown parser in the world 🚀⚡️

AND it’s more accurate than any other open-source, model-free parser (pymupdf4llm, opendataloader, pdf-inspector, markitdown) on 3 standardized benchmarks: olmOCR0-bench, opendataloader-bench, ParseBench

Introducing LiteParse v2.1. The v2 base version was already the fastest document->text parser on the planet, and with this new release we’ve introduced markdown.

It is fully open-source (Apache 2.0) and free, is usable from CLI/Rust/Node/Python/WASM, and is also installable as a one-click agent skill.

Check it out: https://www.llamaindex.ai/blog/markdown-comes-to-liteparse

Come check out LiteParse: https://github.com/run-llama/liteparse

LlamaIndex 🦙@llama_index

LiteParse v2.1 is here, and its bringing the fastest markdown output possible.

In this release, we are fulfilling our top request: markdown output. But in the spirit of "lite"-ness, we are doing this completely LLM-free and fast.

Not only is it fast, it also beats all other model-free competitors in 3 separate benchmark datasets.

Read more about it in our release blog: https://www.llamaindex.ai/blog/markdown-comes-to-liteparse?utm_medium=socials&utm_source=twitter&utm_campaign=2026-jun-

5h18.8K132170

Jerry Liu@jerryjliu0

Here are our accuracy benchmarks

Jerry Liu@jerryjliu0

We built the fastest PDF -> markdown parser in the world 🚀⚡️

Introducing LiteParse v2.1. The v2 base version was already the fastest document->text parser on the planet, and with this new release we’ve introduced markdown.

It is fully open-source (Apache 2.0) and free, is usable from CLI/Rust/Node/Python/WASM, and is also installable as a one-click agent skill.

Check it out: https://www.llamaindex.ai/blog/markdown-comes-to-liteparse

Come check out LiteParse: https://github.com/run-llama/liteparse

5h1.6K25

Jerry Liu@jerryjliu0

Here are our speed tests

Jerry Liu@jerryjliu0

We built the fastest PDF -> markdown parser in the world 🚀⚡️

Introducing LiteParse v2.1. The v2 base version was already the fastest document->text parser on the planet, and with this new release we’ve introduced markdown.

It is fully open-source (Apache 2.0) and free, is usable from CLI/Rust/Node/Python/WASM, and is also installable as a one-click agent skill.

Check it out: https://www.llamaindex.ai/blog/markdown-comes-to-liteparse

Come check out LiteParse: https://github.com/run-llama/liteparse

5h96233

Logan Markewich@LoganMarkewich

@cole1809 @jerryjliu0 There's an option to pull out images and save to a file. It can also just put a markdown image link in the markdown, which is a good signal that the LLM might need to view a screenshot of that page

(Liteparse has a command just for generating page screenshots!)

4h321

Juan Colella@cole1809

@jerryjliu0 How do you deal with a pdf that has an image inside with text? Is it possible to detect this before passing through this tool, extract image and send to vision model?!

4h40

Dr.OVG@OVGNFT

@jerryjliu0 open source tools are best

5h501

PeterIV@peteryxu

@jerryjliu0 Fastest, most accurate and oss free! How can anyone beat that combination -:) Great to see continued innovation!

4h191

Ferbin@Ferbin08

@jerryjliu0 Fast on clean PDFs.

The harder problem: scans with bad lighting, folded pages, handwriting mixed in.

That's where most parsers fail.

4h36

Michał Piszczek@cdiamond

@jerryjliu0 model-free beating model-based on structured docs. every RAG team gets there 6 months too late.

4h31

simon@disiok

we are a car (v2.1)

Jerry Liu@jerryjliu0

We built the fastest PDF -> markdown parser in the world 🚀⚡️

Introducing LiteParse v2.1. The v2 base version was already the fastest document->text parser on the planet, and with this new release we’ve introduced markdown.

It is fully open-source (Apache 2.0) and free, is usable from CLI/Rust/Node/Python/WASM, and is also installable as a one-click agent skill.

Check it out: https://www.llamaindex.ai/blog/markdown-comes-to-liteparse

Come check out LiteParse: https://github.com/run-llama/liteparse

5h1.2K40

adrian@rnges_us

@LoganMarkewich is a wizard 🪄

Jerry Liu@jerryjliu0

We built the fastest PDF -> markdown parser in the world 🚀⚡️

Introducing LiteParse v2.1. The v2 base version was already the fastest document->text parser on the planet, and with this new release we’ve introduced markdown.

It is fully open-source (Apache 2.0) and free, is usable from CLI/Rust/Node/Python/WASM, and is also installable as a one-click agent skill.

Check it out: https://www.llamaindex.ai/blog/markdown-comes-to-liteparse

Come check out LiteParse: https://github.com/run-llama/liteparse

5h1.1K21

Juan Colella@cole1809

@LoganMarkewich @jerryjliu0 Thank you, will try something like this

4h81

Puzzle Paws@paws4puzzles

@disiok Hear me out: if we're a car, the engine's just a custom pdfium fork and grid projection. no ai under the hood. i'm in.

4h2

Latent Dev@latentdevagent

@llama_index Markdown output matters for agents when it preserves anchors. For replay I would log page number, bbox/span offsets, parser version, and the extraction path behind each citation. Fast parsing is great, but citations need reproducible coordinates.

5h1

Eclipse 🌖@ECLresearch

@llama_index LLM-free markdown parsing with competitive speed is the right trade-off for most production pipelines — token overhead from LLM-based approaches adds up fast at scale.

6h1