LlamaIndex 🦙(@llama_index)

The OSS models on OpenDocRouter are slowly powering up! 🚀 We trained additional speculative decod...

LlamaIndex 🦙 (@llama_index) — Mon, 19 Oct 2026 16:00:14 GMT

The OSS models on OpenDocRouter are slowly powering up! 🚀 

We trained additional speculative decoders to help with latency. You should see considerably faster inference times using models like 
@LightOnIO LightOnOCR-3-1B!

We measured tokens per second across single requests (best case) and 64 requests (saturated throughput). TeleOCR sees less speedup at scale since it makes a lot of tiny calls, where decoding latency matters a bit less.

Try these models out today!opendocrouter.aiC

💬9🔄3❤️12👀3060📊10

⚡ Powered by xgo.ing

Source: Table from Simon Ott et al., VL-DocIR, CC BY 4.0 (https://t.co/TUmK5MKeEg).

LlamaIndex 🦙 (@llama_index) — Fri, 16 Oct 2026 16:48:19 GMT

Source: Table from Simon Ott et al., VL-DocIR, CC BY 4.0 (creativecommons.org/licenses/by/4.…).

💬0🔄0❤️4👀958📊1

⚡ Powered by xgo.ing

Doc of the Week: VL-DocIR’s, a new paper on long-document retrieval. Table 2, where the “Data Source...

LlamaIndex 🦙 (@llama_index) — Fri, 16 Oct 2026 16:48:18 GMT

Doc of the Week: VL-DocIR’s, a new paper on long-document retrieval. Table 2, where the “Data Source” header spans Wiki, arXiv, PubMed and SEC and stops one column short of Total.

LlamaParse Agentic Plus kept that span, along with all 35 counts and every label. Try it on your own research paper ->

💬8🔄1❤️13👀3311📊9

⚡ Powered by xgo.ing

There are more document parsing models than ever, and they’re getting harder to tell apart. That’s ...

LlamaIndex 🦙 (@llama_index) — Wed, 14 Oct 2026 16:29:49 GMT

There are more document parsing models than ever, and they’re getting harder to tell apart.

That’s why we built OpenDocRouter. Every model we serve is benchmarked on ParseBench, with latency and cost per page tracked next to accuracy, so you can compare them side by side and choose based on what your documents need.

Failed pages are free and new accounts start with $5 in credit. Try it at opendocrouter.ai!

Your browser does not support the video tag.

🔗 View on Twitter

💬18🔄2❤️24👀4231📊20

⚡ Powered by xgo.ing

OpenDocRouter makes it dead-simple to start parsing your documents across on ton of frontier Models ...

LlamaIndex 🦙 (@llama_index) — Tue, 13 Oct 2026 19:00:27 GMT

OpenDocRouter makes it dead-simple to start parsing your documents across on ton of frontier Models (gemini, luna) and latest OSS models (LightOnOCR-3, MinerU2.5-pro). Parse your first document in 60 seconds!

opendocrouter.ai

Your browser does not support the video tag.

🔗 View on Twitter

💬11🔄2❤️17👀6487📊12

⚡ Powered by xgo.ing

RSVP here! https://t.co/P8mrpYLzXB

LlamaIndex 🦙 (@llama_index) — Tue, 13 Oct 2026 17:14:47 GMT

RSVP here!
luma.com/llamai-68ea

💬0🔄1❤️2👀1615📊1

⚡ Powered by xgo.ing

Doc Chai is real, and on Thursday you can drink it ☕ We’re throwing a launch party for our first ...

LlamaIndex 🦙 (@llama_index) — Tue, 13 Oct 2026 17:14:28 GMT

Doc Chai is real, and on Thursday you can drink it ☕  

We’re throwing a launch party for our first drink with @useCorgi Cafe: an iced toasted brown sugar chai!  

For many of us on the team, chai is what home tastes like.  

Come say hi, 600+ people have RSVP'ed and the first 100 drinks are on us. We'll have a live DJ to wake everyone up as well :)  

8-11 AM, Corgi Cafe, FiDi 🦙  

RSVP link in the comments below.

💬3🔄1❤️13👀5925📊5

⚡ Powered by xgo.ing

Three events, one award, a very respectable 4th place in hot sauce. Our SF Tech Week: → AI Demo Ni...

LlamaIndex 🦙 (@llama_index) — Mon, 12 Oct 2026 16:30:21 GMT

Three events, one award, a very respectable 4th place in hot sauce.

Our SF Tech Week:

→ AI Demo Night w/ @ClickHouse + @artie_labs: @imaanxsultan won best demo

→ @matissemlk took 4th in the Blazel hot sauce competition

→ Co-hosted the biggest night of SF Tech Week with @SpaceX, @Lemma_, @Daytonaio, @Tiny_Fish, and more! Agent Economy Gala: 500+ founders + investors in suits and gowns

Thanks to everyone who joined us for a very busy, but fun SF Tech Week :)

💬6🔄3❤️12👀6402📊7

⚡ Powered by xgo.ing

Source linked here: https://t.co/Q3FtQcyuaW

LlamaIndex 🦙 (@llama_index) — Fri, 09 Oct 2026 16:24:05 GMT

Source linked here: s25.q4cdn.com/621799436/file…

💬0🔄0❤️0👀1392

⚡ Powered by xgo.ing

What happens when a table contains tables? Usually, a mess. With LlamaParse, it comes out clean. D...

LlamaIndex 🦙 (@llama_index) — Fri, 09 Oct 2026 16:21:50 GMT

What happens when a table contains tables?

Usually, a mess. With LlamaParse, it comes out clean.

Doc of the Week: Micron’s latest earnings deck. Even with two business units and same row names, LlamaParse keeps all 18 values under the right header.

Try it out on your own earnings deck: llamaindex.ai/llamaparse

💬13🔄5❤️30👀5488📊18

⚡ Powered by xgo.ing

Markdown is all you need. (Mostly.) A parser can get every word on the page right and still lose ...

LlamaIndex 🦙 (@llama_index) — Thu, 08 Oct 2026 16:49:14 GMT

Markdown is all you need. 

(Mostly.)

A parser can get every word on the page right and still lose which column a number belongs to. Then your model has to guess.

Markdown keeps headings, lists, and tables intact, and it stays readable when you're debugging a bad answer. For tables with merged headers, we switch to HTML.

Read our breakdown on why it's our default output for parsing below! ⬇️

💬19🔄7❤️114👀12959📊41

⚡ Powered by xgo.ing

Today we're announcing OpenDocRouter: every model for document parsing under one API. There are ~4,...

LlamaIndex 🦙 (@llama_index) — Wed, 07 Oct 2026 16:01:35 GMT

Today we're announcing OpenDocRouter: every model for document parsing under one API.

There are ~4,000 OCR models on Hugging Face, and the frontier labs ship a new one nearly every month. Whatever's best for your docs today won't be by Q1.

So stop asking which model to use for doc parsing. Ask how fast you can switch.

✅ Switch models in one line: same request, same markdown output, no new prompts or integrations
✅ Any model you want: 10 frontier and open-source models at launch, including Claude Opus 5.5, Gemini 3.8 Flash, GPT-6 Luna, MinerU2.5-Pro and PaddleOCR-VL-1.6
✅ Choose with receipts: every model scored on ParseBench for quality and cost
✅ Traceable output from any model: "layout: true" adds grounded bounding boxes and the same layout classes, even for models that don't support it natively
✅ Pay only for what works: per-token pricing, failed pages never charged, top up from $25

The spread is the point: $0.86 to $48.82 per 1,000 pages, depending on what your documents actually need.

Live now → opendocrouter.ai

Your browser does not support the video tag.

🔗 View on Twitter

💬34🔄37❤️259👀66059📊95

⚡ Powered by xgo.ing

Request a spot here! https://t.co/7DlvDslr49

LlamaIndex 🦙 (@llama_index) — Tue, 06 Oct 2026 17:12:04 GMT

Request a spot here! partiful.com/e/ynm6aOPAPq4Q…

💬0🔄1❤️1👀1177📊1

⚡ Powered by xgo.ing

The biggest night of SF Tech Week is black tie. 120+ agent founders, 50 investors, and live jazz at...

LlamaIndex 🦙 (@llama_index) — Tue, 06 Oct 2026 17:11:17 GMT

The biggest night of SF Tech Week is black tie.

120+ agent founders, 50 investors, and live jazz at the Redwood Room on Friday for the Agent Economy Gala. Invite only.

We're excited to be partnering with @SpaceX, @Daytonaio, @lemma_, @Tiny_Fish, @thelinqapp, @Obvious & more. Link to request an invite in below or comment what you're building!

💬5🔄4❤️14👀4000📊7

⚡ Powered by xgo.ing

Read our take here: https://t.co/xJsRJx7VGL

LlamaIndex 🦙 (@llama_index) — Mon, 05 Oct 2026 18:48:44 GMT

Read our take here: llamaindex.ai/blog/ocr-is-de…

💬0🔄5❤️27👀3678📊11

⚡ Powered by xgo.ing

OCR is dead 🪦 Long live agentic OCR! Traditional OCR makes one pass and hands back whatever text i...

LlamaIndex 🦙 (@llama_index) — Mon, 05 Oct 2026 18:47:44 GMT

OCR is dead 🪦 Long live agentic OCR!

Traditional OCR makes one pass and hands back whatever text it got. Tables get flattened, charts disappear, and multi-column layouts come out scrambled. Nothing checks the output.

Agentic OCR treats parsing as a loop instead:

✅ layout-aware reading order
✅ smart routing of hard elements to the right model
✅ multi-pass verification and self-correction
✅ multimodal parsing of charts, images, and complex tables

Logan Markewich, LlamaIndex's Head of Open Source, wrote about what that shift looks like in practice and where it's still struggling. Link to the breakdown in the comments below.

💬27🔄73❤️667👀60540📊224

⚡ Powered by xgo.ing

The BEA's new Q2 GDP-by-state was released, but the key data is entirely in an infographic, without ...

LlamaIndex 🦙 (@llama_index) — Fri, 02 Oct 2026 17:35:04 GMT

The BEA's new Q2 GDP-by-state was released, but the key data is entirely in an infographic, without a text-layer.

LlamaParse's agentic parsing tier understands documents like this by default, and faithfully records data, turning the page into a 52-row table.

Source: bea.gov/sites/default/…

💬8🔄5❤️10👀4441📊9

⚡ Powered by xgo.ing

How many people show up on a Tuesday night to talk about document processing for AI agents? 🤔 In N...

LlamaIndex 🦙 (@llama_index) — Wed, 30 Sep 2026 18:32:22 GMT

How many people show up on a Tuesday night to talk about document processing for AI agents? 🤔

In New York, enough to fill the room and start a waitlist. In San Francisco, almost 600.

We were in both cities on the same night to talk about one problem: agents don't read documents the way people do. Look at an invoice and you can tell the monthly rate from the total in a second. An agent often gets the words without the layout, so it has to guess.

In New York, we partnered with @usekernel on @braintrust's Agent Builders Night. In San Francisco, we partnered with @WorkOS on @daytonaio's AI Builders event. Thanks to them for having us, and to Abrar and Yong from our team for speaking!

💬9🔄7❤️20👀4359📊11

⚡ Powered by xgo.ing

Document parsing requires a lot of on-the-fly decision making. We explored the early approaches to ...

LlamaIndex 🦙 (@llama_index) — Mon, 28 Sep 2026 21:00:04 GMT

Document parsing requires a lot of on-the-fly decision making.

We explored the early approaches to Jev and Jev-like models on several documents tasks like orientation detection, language detection, routing, and more.
youtube.com/watch?v=MLZ1dU…

💬4🔄6❤️70👀31729📊17

⚡ Powered by xgo.ing

Parsing forms still trips up the latest frontier VLMs. A form isn't text on a page. It's a set of f...

LlamaIndex 🦙 (@llama_index) — Mon, 28 Sep 2026 19:25:38 GMT

Parsing forms still trips up the latest frontier VLMs.

A form isn't text on a page. It's a set of fields, grouped into sections, each tied to a specific box. That's why forms need purpose-built parsing, not a bigger general model:

✅️ Detect every field, not just the obvious ones
✅️ Keep the hierarchy of sections and fields
✅️ Tie every value to the exact box it came from
✅️ Understand handwriting and checkmarks

Our latest  blogpost breaks down where VLMs fail on real W-2s, 1040s, W-9s and scanned W-4s, and a custom cookbook for LlamaParse to handle them at a fraction of the cost 👇 
llamaindex.ai/blog/why-vlms-…y

💬8🔄2❤️28👀12126📊11

⚡ Powered by xgo.ing

A blank cell can change the meaning of a forecast. Four times a year, the Fed's 18 top policymakers...

LlamaIndex 🦙 (@llama_index) — Fri, 25 Sep 2026 15:12:36 GMT

A blank cell can change the meaning of a forecast.

Four times a year, the Fed's 18 top policymakers each put their forecasts on paper: where growth, jobs, inflation, and interest rates are headed. This is September 2026's edition, home of the "dot plot" that markets treat as the Fed tipping its hand. This release is the closest thing to the Fed saying what it plans to do. Most analysts will want to throw this documents to an AI agent, but this messy doc is dense: full of complex tables and charts that hold valuable context. All things that frequently trip up raw LLM APIs.

The Fed’s September 2026 projections table includes a 2029 column, but its June comparison row leaves that cell empty. This is a common but silent failure point for document parsers.

We parsed page 2 with LlamaParse and checked the displayed GDP median excerpt against the original PDF. All nine numbers matched and the data stays aligned in the returned HTML.

Try LlamaParse on a table where headers and missing cells matter. cloud.llamaindex.ai/signup

Source: federalreserve.gov/monetarypolicy…

💬4🔄3❤️17👀9566📊7

⚡ Powered by xgo.ing

confidence scores only matter if they help you decide what to automate. for document extraction, th...

LlamaIndex 🦙 (@llama_index) — Thu, 24 Sep 2026 18:41:23 GMT

confidence scores only matter if they help you decide what to automate.

for document extraction, that usually means knowing how much work you can safely accept at a given precision target. in our latest post, we look at confidence scoring through that lens, including:

✅️  confidence cutoffs
✅️  precision vs. recall
✅️  score coverage
✅️  score granularity
✅️  human review volume

using ExtractBench, we compare how different extraction systems perform after confidence filtering. at a 97% precision target, LlamaParse Agentic Plus reached 66.48% recall on expected fields after filtering.

the useful part of a confidence score isn’t the number itself. it’s whether you can use it to control automation and review in production.

👉️ read the full post:llamaindex.ai/blog/what-make…f

Your browser does not support the video tag.

🔗 View on Twitter

💬7🔄6❤️24👀18919📊12

⚡ Powered by xgo.ing

The fastest PDF-to-Markdown parser just got faster. ⚡️ With LiteParse v2.14.6, text-based PDFs pars...

LlamaIndex 🦙 (@llama_index) — Tue, 22 Sep 2026 17:03:02 GMT

The fastest PDF-to-Markdown parser just got faster. ⚡️

With LiteParse v2.14.6, text-based PDFs parse about 25% faster.

Across realistic documents, LiteParse processed pages at 2.8ms/page and 1.5× faster than the next-fastest local parser.

LiteParse is open source and runs locally in Python, Node.js, Rust, or directly in the browser.

Grab v2.14.6 → github.com/run-llama/lite…
Bench Docs →  huggingface.co/datasets/llama…

💬13🔄1❤️69👀30971📊31

⚡ Powered by xgo.ing

Grounded Confidence is here for Extract! 🦙 When your agents and workflows depend on extracted data...

LlamaIndex 🦙 (@llama_index) — Mon, 21 Sep 2026 17:03:20 GMT

Grounded Confidence is here for Extract! 🦙

When your agents and workflows depend on extracted data, you need to know how accurate that data is.

We’ve added confidence scores to give you a better read on extraction accuracy, field by field. Use them to decide which results your app can accept automatically and which need human review.

Available on Cost Effective, Agentic, and Agentic Plus.

Try it out on your docs ->developers.llamaindex.ai/llamaparse/ext…t

Your browser does not support the video tag.

🔗 View on Twitter

💬10🔄2❤️41👀17226📊15

⚡ Powered by xgo.ing

Introducing the first in our Parsed by LlamaParse series. We're outlining the stakes of 'getting par...

LlamaIndex 🦙 (@llama_index) — Fri, 18 Sep 2026 14:46:03 GMT

Introducing the first in our Parsed by LlamaParse series. We're outlining the stakes of 'getting parsing wrong' in consequential documents. 

We’re starting with the U.S. Energy Information Administration’s September 2026 Short-Term Energy Outlook, a dense government report covering energy supply, demand, prices, and forecasts.

Table 7a alone packs multiple years, quarters, row hierarchies, units, and footnotes into one electricity-industry table.

Take 1,186.

Parsed correctly, it means electricity sales to ultimate customers in Q3 2026, measured in billion kilowatthours.

Parsed incorrectly, it could be assigned to the wrong quarter, metric, or unit , which means the error can flow straight into a dashboard, forecast, alert, or AI application.

And the values aren’t the only thing that matters. Footnotes define the data too: “small-scale solar,” for example, refers to systems under one megawatt, not solar generation overall.

LlamaParse preserves the structure and context that make document data usable downstream: whether you’re populating a database, updating a dashboard, running forecasting workflows, or building an AI app over complex documents.

Source doc here: eia.gov/outlooks/steo/

💬17🔄11❤️36👀5162📊22

⚡ Powered by xgo.ing

It was great to be at Connected Stack last week with founders and builders working on what’s next in...

LlamaIndex 🦙 (@llama_index) — Thu, 17 Sep 2026 19:49:55 GMT

It was great to be at Connected Stack last week with founders and builders working on what’s next in enterprise AI.

@jerryjliu0 joined the Founder Flash Talks to talk about a problem every enterprise agent eventually runs into: messy, complex documents that general-purpose models struggle to read.

Agents are the new knowledge workers, and we are building the document infrastructure for agents. 

Thanks @trueventures and @GreylockVC for having us! 📸

💬9🔄5❤️20👀4377📊11

⚡ Powered by xgo.ing

Ready to learn how to build document processing pipelines for insurance agents? Join LlamaIndex So...

LlamaIndex 🦙 (@llama_index) — Wed, 16 Sep 2026 16:00:03 GMT

Ready to learn how to build document processing pipelines for insurance agents? 

Join LlamaIndex Solutions Architect Abrar Mahi for a webinar on turning insurance documents into structured data for underwriting, policy review, and claims workflows.

Using LlamaParse and Extract, we’ll cover:

✅️ Parsing complex accord forms, policy documents and other popular document types in the insurance industry. 
✅️ Extracting policy details, property information, and claims history into a defined schema.
✅️ Verifying extracted values with citations and bounding boxes.
✅️ Combining confidence scores with validation rules to decide what proceeds automatically and what needs human review.

We’ll bring these steps together in a working pipeline you can adapt to your own workflows.

Register: watch.getcontrast.io/register/llama…

💬5🔄4❤️17👀4524📊7

⚡ Powered by xgo.ing

keeping our SDKs up to date can be a pain. every API change brings another round of updates, tests, ...

LlamaIndex 🦙 (@llama_index) — Tue, 15 Sep 2026 16:00:02 GMT

keeping our SDKs up to date can be a pain. every API change brings another round of updates, tests, and releases. and each SDK should still feel like it was written by someone who actually uses the language.

Stainless helped us do that for LlamaParse. It also pushed us to improve the API itself. inconsistent names and schemas become harder to ignore when they show up in the code developers use.

with the Stainless team joining Anthropic, George He and Yong Park wrote about what worked, what we learned, and why changing SDK generators takes more care than you might expect. -> llamaindex.ai/blog/au-revoir…

thanks to the Stainless team for saving us a lot of SDK work. 👋

au revoir, Stainless

💬3🔄3❤️16👀11023📊6

⚡ Powered by xgo.ing

just-in-time OCR is all the rage. most pipelines parse every page before anyone asks a question. fo...

LlamaIndex 🦙 (@llama_index) — Mon, 14 Sep 2026 16:00:11 GMT

just-in-time OCR is all the rage.

most pipelines parse every page before anyone asks a question. for an agent working through an ad-hoc data room, that's slow, expensive, and most of those pages never get read.

the better pattern is just-in-time OCR in two passes:

✅️  LiteParse (free, OSS, Rust, 50+ formats) does a fast layout-aware first pass: spatial text, bounding boxes, headings, tables, and a per-page complexity flag. a full data room in 32 seconds.

✅️ LlamaParse zooms in on only the pages that need it, by page number, and returns cell-level tables, bounding boxes, and confidence scores. the rest fills in the background.

pypdf and pdftotext can't do the first pass well. parsing everything up front can't do it cheaply. two passes gets you both.

full breakdown with numbers: llamaindex.ai/blog/just-in-t…

💬21🔄11❤️84👀26684📊36

⚡ Powered by xgo.ing

confidence scores in LlamaParse just got an upgrade 🦸‍♀️ our new high-effort mode provides granul...

LlamaIndex 🦙 (@llama_index) — Fri, 11 Sep 2026 13:46:01 GMT

confidence scores in LlamaParse just got an upgrade 🦸‍♀️ 

our new high-effort mode provides granular page-level scores with text explanations to help you intimately understand parsing quality of your documents. we even refer back to the original document for an extra check when generating the score. 

use high-effort only when you need it, at 5 additional credits per page.

try it on your docs:developers.llamaindex.ai/llamaparse/par…h

💬13🔄6❤️35👀12289📊17

⚡ Powered by xgo.ing

Last week we brought LlamaParse to Claude. Now it’s available in ChatGPT, too 🦙 ChatGPT can alrea...

LlamaIndex 🦙 (@llama_index) — Wed, 09 Sep 2026 16:58:16 GMT

Last week we brought LlamaParse to Claude. Now it’s available in ChatGPT, too 🦙 

ChatGPT can already read PDFs. But document work gets harder when the information you need is buried in messy scans, dense tables, spreadsheets, and charts. Strike the right balance between burning tokens and accuracy, with a specialized API for document understanding.

Turn complex documents into clean, structured context and let Astra run its magic. Use our connector to:

✅ Parse documents into clean Markdown, JSON, or HTML. 
✅ Extract specific fields into a schema you define.
✅ Search across document collections to find the information you need. 
✅ Classify files and split documents into logical sections.

Live in the ChatGPT plugin directory now! →chatgpt.com/plugins/plugin…5

Your browser does not support the video tag.

🔗 View on Twitter

💬5🔄4❤️29👀18168📊11

⚡ Powered by xgo.ing

Does paying 5x more per page actually get you better document extraction? We ran the data to find ou...

LlamaIndex 🦙 (@llama_index) — Fri, 04 Sep 2026 16:43:38 GMT

Does paying 5x more per page actually get you better document extraction? We ran the data to find out.

We evaluated 14 frontier systems across 370 enterprise documents, then plotted their accuracy against cost per page. The biggest takeaway? Higher cost does not equal better extraction.

• Agentic Plus hit the highest accuracy overall at <1/3 the cost of the runner-up.
• Agentic & Cost Effective routinely beat systems costing several times more per page.

More expensive doesn’t necessarily mean more accurate. Now we have the data to show it.

Try Extract on your own documents with 10,000 free credits when you sign up for LlamaParse → cloud.llamaindex.ai/?utm_medium=so…

💬8🔄4❤️26👀15285📊12

⚡ Powered by xgo.ing

Document extraction just got a new gear. ⚡️ Introducing Turbo mode for Extract, our fastest way yet...

LlamaIndex 🦙 (@llama_index) — Thu, 03 Sep 2026 16:49:39 GMT

Document extraction just got a new gear. ⚡️

Introducing Turbo mode for Extract, our fastest way yet to pull structured data from documents.

It runs roughly 4× faster than our Cost Effective Tier at comparable accuracy, with a median latency of 3.7 seconds a page. Turbo processes pages in parallel, so latency stays nearly flat as document size grows. 

Available in beta today.

Read the full breakdown → llamaindex.ai/blog/introduci…

💬9🔄10❤️46👀23403📊20

⚡ Powered by xgo.ing

ExtractBench is now live on @Kaggle. It tests schema-guided document extraction on the documents mo...

LlamaIndex 🦙 (@llama_index) — Wed, 02 Sep 2026 16:01:20 GMT

ExtractBench is now live on @Kaggle.

It tests schema-guided document extraction on the documents most likely to break downstream agents and workflows, including long record lists, noisy scans, handwriting, and complex tables.

The benchmark covers 370 enterprise documents across 8 business domains and 67 document types. See how LlamaParse, Codex, Claude Code, and more stack up.

Read the full story →llamaindex.ai/blog/llamainde…

💬5🔄6❤️32👀10166📊10

⚡ Powered by xgo.ing

LlamaParse is officially a verified connector on Claude! 🦙 Claude is perfectly fine for the occasi...

LlamaIndex 🦙 (@llama_index) — Mon, 31 Aug 2026 16:11:46 GMT

LlamaParse is officially a verified connector on Claude! 🦙

Claude is perfectly fine for the occasional text-only PDF. But most of the context your agent needs lives in messy PDFs, spreadsheets, scanned forms, dense tables, and charts. 

Feeding those docs straight to the model burns through tokens and can lead to hallucinations when layouts get scrambled, chart values are missed, or numbers are pulled from the wrong column. 

LlamaParse converts those documents into clean, structured context Claude can work with. With the connector you can

✅️ Parse documents into clean Markdown, JSON, or HTML
✅️ Extract specific fields directly into your schema
✅️ Search across document collections with filesystem-style tools
✅️ Classify documents or split them into predefined sections

Live in the Claude Connectors Directory now !
claude.ai/directory/llam…4

Your browser does not support the video tag.

🔗 View on Twitter

💬7🔄4❤️38👀13535📊14

⚡ Powered by xgo.ing

Static embeddings offer unmatched throughput, but they also suffer from accuracy loss compared to tr...

LlamaIndex 🦙 (@llama_index) — Fri, 28 Aug 2026 15:02:11 GMT

Static embeddings offer unmatched throughput, but they also suffer from accuracy loss compared to traditional embedding models. Can we make them better for retrieval?

We tried several things:
✅️  raw maxsim scoring on the per-token embeddings
✅️  training a small adapter model
✅️  changing the distillation training target and teacher

While none of these saw the results we wanted, it's an excellent dive into static embeddings and how they do (and don't) work!

Blog: llamaindex.ai/blog/exploring…

💬5🔄7❤️31👀13492📊12

⚡ Powered by xgo.ing

Most extraction tools treat spreadsheets like PDFs. They flatten the file into text or markdown, the...

LlamaIndex 🦙 (@llama_index) — Thu, 27 Aug 2026 16:26:20 GMT

Most extraction tools treat spreadsheets like PDFs. They flatten the file into text or markdown, then ask a model to infer the original structure.

But spreadsheets depend on structure. Headers, formulas, merged cells, and hidden rows give every value its context. Strip that away, and you map the right number to the wrong metric or period.

That's why we built native spreadsheet extraction into the LlamaParse platform.

Instead of flattening your workbook to text, it reads the raw cells directly and maps the data to your schema.

Available today in beta on the agentic_plus tier. Give it a spin on your messiest .xlsx, .xls, or .csv files.

Docs: developers.llamaindex.ai/llamaparse/ext…

💬3🔄11❤️57👀15342📊17

⚡ Powered by xgo.ing

Our 2nd founder dinner in SF co-hosted by @jerryjliu0 and @GuangyuRobert at @Fundamental - the team ...

LlamaIndex 🦙 (@llama_index) — Wed, 26 Aug 2026 15:00:08 GMT

Our 2nd founder dinner in SF co-hosted by @jerryjliu0 and @GuangyuRobert at @Fundamental - the team behind @tryshortcutai Talking about existing moats in the AI era.

Frontier labs are moving past model APIs into vertical agents - ChatGPT Health, Claude for Legal.

So where does the moat sit now?

✅ Agent engineering
✅ Infra optimization
✅ Domain evals and data
✅ Workflow expertise
✅ GTM and brand

If you're a founder or CTO shipping agents in production and want to know what other teams are doing to maintain their moat. Request a seat 👉️luma.com/llamai-8hryz

💬1🔄1❤️7👀4352📊2

⚡ Powered by xgo.ing

Every extraction API demos well on a clean invoice. But what about the scanned form, the nested tabl...

LlamaIndex 🦙 (@llama_index) — Tue, 25 Aug 2026 17:30:23 GMT

Every extraction API demos well on a clean invoice. But what about the scanned form, the nested table, the 40-page financial report with merged headers?

We tested 14 frontier systems to find out. ExtractBench evaluates schema-guided extraction across 370 enterprise documents, 67 document types, and 4,800+ pages.

Join Simon Suo, CTO and co-founder of LlamaIndex, for a technical walkthrough of the results: what makes extraction hard, where existing benchmarks miss it, and how cost trades off against accuracy across VLMs, coding agents, and specialized APIs.

Wednesday, Aug 26 · 9:00 AM PT / 12:00 PM ET 

Save your seat 👉lnkd.in/gMbFNHPg8

💬6🔄2❤️14👀7873📊7

⚡ Powered by xgo.ing

Our 2nd founder dinner in SF co-hosted by @jerryjliu0 and @GuangyuRobert at @Fundamental - the team ...

LlamaIndex 🦙 (@llama_index) — Mon, 24 Aug 2026 20:10:45 GMT

Our 2nd founder dinner in SF co-hosted by @jerryjliu0 and @GuangyuRobert at @Fundamental - the team behind @tryshortcutai Talking about existing moats in the AI era.

Frontier labs are moving past model APIs into vertical agents - ChatGPT Health, Claude for Legal.

So where does the moat sit now?

✅ Agent engineering
✅ Infra optimization
✅ Domain evals and data
✅ Workflow expertise
✅ GTM and brand

If you're a founder or CTO shipping agents in production and want to know what other teams are doing to maintain their moat. Request a seat 👉️luma.com/llamai-8hryz

💬1🔄2❤️8👀3301📊2

⚡ Powered by xgo.ing

Our CEO @jerryjliu0 is speaking at @CoreWeave Fully Connected 2026 in SF. 🎤 "Automating Document W...

LlamaIndex 🦙 (@llama_index) — Fri, 21 Aug 2026 13:56:08 GMT

Our CEO @jerryjliu0 is speaking at @CoreWeave Fully Connected 2026 in SF. 🎤

"Automating Document Work with Long-Horizon AI Agents" Sept 30, Moscone Center.

Come see how long-horizon agents are taking on real document workflows in production. See you there!

Checkout the full lineup �coreweave.com/fully-connecte…TM

💬6🔄6❤️25👀4836📊9

⚡ Powered by xgo.ing

Most parsers treat tracked changes as noise. Result: a deleted clause comes back as live text, and y...

LlamaIndex 🦙 (@llama_index) — Tue, 18 Aug 2026 15:31:37 GMT

Most parsers treat tracked changes as noise. Result: a deleted clause comes back as live text, and your pipeline reads a document that says the opposite of what it means.

LlamaParse now handles revision tracking. You get clean markdown of the document's final state, plus every revision (edits, deletions, comments) as structured data with author, content, and location. 

Contracts, regulatory submissions, policy drafts: if your documents go through redline rounds, this is for you.

Docs 👉️ :developers.llamaindex.ai/llamaparse/par…3

Sign Up:cloud.llamaindex.ai/?utm_medium=so…y

Your browser does not support the video tag.

🔗 View on Twitter

💬9🔄5❤️33👀13961📊14

⚡ Powered by xgo.ing

Most document extraction APIs can't tell you where a value came from. For ExtractBench, we scored g...

LlamaIndex 🦙 (@llama_index) — Mon, 17 Aug 2026 15:59:35 GMT

Most document extraction APIs can't tell you where a value came from.

For ExtractBench, we scored grounding strictly: a field only counts if the value AND its citation are correct, word-level box at IoU 0.5. A perfect box around a wrong value earns nothing.

Results: VLMs and coding agents return no evidence at all — zero at both levels. Among systems that do return boxes, the best word-level F1 is still under 50%. And grounding collapses with length: one specialized API goes from 61.7% page-level on short docs to 0.0% on long ones.

LlamaExtract Agentic Plus leads at both levels — 84.9% page-level, 46.4% word-level — and holds at 87.1% on long documents where others hit zero.

Every extracted value should come with receipts. ExtractBench now gives the field a baseline to track it 👇

Blog:lnkd.in/gNm97fXp6
Paper:lnkd.in/euAfScWxF

Your browser does not support the video tag.

🔗 View on Twitter

💬11🔄5❤️36👀22008📊16

⚡ Powered by xgo.ing

Get ready to break racks and agents. If you're in SF, next Thursday we're co-hosting a billiards n...

LlamaIndex 🦙 (@llama_index) — Fri, 14 Aug 2026 18:30:01 GMT

Get ready to break racks and agents. 

If you're in SF, next Thursday we're co-hosting a billiards night for AI engineers and founders with Judgment Labs , Modal, and ClickHouse.

Limited spots available. RSVPs are being approved now 👇️
luma.com/bj9o4jzdc

💬8🔄2❤️15👀5047📊9

⚡ Powered by xgo.ing

🦙☀️ It’s a hot Llama summer, and it pays to be a PRO at document parsing. For a limited time, upgra...

LlamaIndex 🦙 (@llama_index) — Thu, 13 Aug 2026 18:18:24 GMT

🦙☀️ It’s a hot Llama summer, and it pays to be a PRO at document parsing. For a limited time, upgrade to the LlamaParse Pro plan and we'll triple your credits:

400K → 1.2M credits in your first month. That's $1,000 (~80k pages) of parsing, free.

No code, applied automatically. Upgrade today 👇️ llamaindex.ai/pricingZP

💬8🔄2❤️25👀334145📊11

⚡ Powered by xgo.ing

Every document extraction system has a perception blind spot. We mapped them. For ExtractBench, we ...

LlamaIndex 🦙 (@llama_index) — Thu, 13 Aug 2026 14:05:11 GMT

Every document extraction system has a perception blind spot. We mapped them.

For ExtractBench, we tested 14 systems on documents that weren't born digital: 1950s regulatory filings, hand-filled tax forms, and pages degraded with fax thresholding, photocopier tone curves, sensor noise, and phone-camera capture.

The failures don't overlap. 

💡 Codex reads scans and handwriting above 93%, then drops to ~80% on rotated or image-only pages. 
💡 Specialized APIs are the exact inverse: fine on rotation and handwriting, 81% on scans
💡 Gemini 3.5 Flash falls from 88.6% to 71.1% the moment a page is scanned.

You benchmark on clean PDFs. Production sends you a shadowed photocopy from 1953.

Our new Extract Tier, Agentic Plus, was the only system with no blind spot: 95.9% / 93.9% / 93.8% across rotated, scanned, and handwritten, a 2-point spread where others swing 10+.

Learn more about ExtractBench 👇️ 

Bllnkd.in/gNm97fXpceB6
Paplnkd.in/euAfScWxkQpF

Your browser does not support the video tag.

🔗 View on Twitter

💬5🔄7❤️22👀11729📊10

⚡ Powered by xgo.ing

The most dangerous document extraction failure isn't a wrong value. It's a missing row that looks li...

LlamaIndex 🦙 (@llama_index) — Wed, 12 Aug 2026 16:20:34 GMT

The most dangerous document extraction failure isn't a wrong value. It's a missing row that looks like nothing is wrong.

We released ExtractBench yesterday: 370 enterprise docs, 14 systems. The hardest test: long-list completeness. An unclaimed-property list with 26,725 rows. A creditor matrix with 8,624 records. A 13F with 3,063 holdings.

Frontier VLMs don't misread these docs, they abandon them. Precision stays high, recall collapses: 8.9–35.8% F1 on the longest documents. Every row they return looks correct, so spot checks pass while most of the document never came back.

Our new Extract tier, Agentic Plus, processes long docs iteratively instead of one pass: 96.1% F1 on long-list tasks, and the only system that holds flat as docs get longer.

Learn more about ExtractBench below👇

Blog:llamaindex.ai/blog/introduci…Y
Paper:arxiv.org/pdf/2607.29677u

Your browser does not support the video tag.

🔗 View on Twitter

💬6🔄5❤️36👀5787📊13

⚡ Powered by xgo.ing

Introducing 𝗘𝘅𝘁𝗿𝗮𝗰𝘁𝗕𝗲𝗻𝗰𝗵: the most comprehensive benchmark for information extraction fr...

LlamaIndex 🦙 (@llama_index) — Tue, 11 Aug 2026 13:00:51 GMT

Introducing 𝗘𝘅𝘁𝗿𝗮𝗰𝘁𝗕𝗲𝗻𝗰𝗵: the most comprehensive benchmark for information extraction from complex enterprise documents. Our applied research team tested:

14 systems — frontier VLMs, coding agents, extraction APIs — on 370 enterprise docs, 4,869 pages, 67 doc types. Zero LLM judges, fully deterministic.

Biggest finding: past 50 pages, commercial VLMs collapse below 35% recall. Precision stays high, but they silently drop most of the table rows.

What is your extraction agent missing? Run ExtractBench to see todllamaindex.ai/blog/introduci…o/5UE8frCgithub.com/run-llama/Extr…o/27vJ36bL54
Hhuggingface.co/datasets/llama…o/oeGlwPjcQh

Your browser does not support the video tag.

🔗 View on Twitter

💬17🔄11❤️80👀32525📊30

⚡ Powered by xgo.ing

SF builders, it's time to ditch your agents and head out on a 🏃 Join us for a run tomorrow with t...

LlamaIndex 🦙 (@llama_index) — Mon, 10 Aug 2026 18:40:26 GMT

SF builders, it's time to ditch your agents and head out on a 🏃 

Join us for a run tomorrow with the Lore community at 6pm at Sue Bierman Park (across from the Ferry Building) for a casual 5k down the Embarcadero. All paces welcome, founders and startup teams especially👇luma.com/4iiiprxj?utm_m…6g

💬6🔄2❤️12👀5010📊7

LlamaIndex 🦙(@llama_index)

The OSS models on OpenDocRouter are slowly powering up! 🚀 We trained additional speculative decod...

Source: Table from Simon Ott et al., VL-DocIR, CC BY 4.0 (https://t.co/TUmK5MKeEg).

Doc of the Week: VL-DocIR’s, a new paper on long-document retrieval. Table 2, where the “Data Source...

There are more document parsing models than ever, and they’re getting harder to tell apart. That’s ...

OpenDocRouter makes it dead-simple to start parsing your documents across on ton of frontier Models ...

RSVP here! https://t.co/P8mrpYLzXB

Doc Chai is real, and on Thursday you can drink it ☕ We’re throwing a launch party for our first ...

Three events, one award, a very respectable 4th place in hot sauce. Our SF Tech Week: → AI Demo Ni...

Source linked here: https://t.co/Q3FtQcyuaW

What happens when a table contains tables? Usually, a mess. With LlamaParse, it comes out clean. D...

Read more here: https://t.co/SlFFt35S1S

Markdown is all you need. (Mostly.) A parser can get every word on the page right and still lose ...

Today we're announcing OpenDocRouter: every model for document parsing under one API. There are ~4,...

Request a spot here! https://t.co/7DlvDslr49

The biggest night of SF Tech Week is black tie. 120+ agent founders, 50 investors, and live jazz at...

Read our take here: https://t.co/xJsRJx7VGL

OCR is dead 🪦 Long live agentic OCR! Traditional OCR makes one pass and hands back whatever text i...

The BEA's new Q2 GDP-by-state was released, but the key data is entirely in an infographic, without ...

How many people show up on a Tuesday night to talk about document processing for AI agents? 🤔 In N...

Document parsing requires a lot of on-the-fly decision making. We explored the early approaches to ...

Parsing forms still trips up the latest frontier VLMs. A form isn't text on a page. It's a set of f...

A blank cell can change the meaning of a forecast. Four times a year, the Fed's 18 top policymakers...

confidence scores only matter if they help you decide what to automate. for document extraction, th...

The fastest PDF-to-Markdown parser just got faster. ⚡️ With LiteParse v2.14.6, text-based PDFs pars...

Grounded Confidence is here for Extract! 🦙 When your agents and workflows depend on extracted data...

Introducing the first in our Parsed by LlamaParse series. We're outlining the stakes of 'getting par...

It was great to be at Connected Stack last week with founders and builders working on what’s next in...

Ready to learn how to build document processing pipelines for insurance agents? Join LlamaIndex So...

keeping our SDKs up to date can be a pain. every API change brings another round of updates, tests, ...

just-in-time OCR is all the rage. most pipelines parse every page before anyone asks a question. fo...

confidence scores in LlamaParse just got an upgrade 🦸‍♀️ our new high-effort mode provides granul...

Last week we brought LlamaParse to Claude. Now it’s available in ChatGPT, too 🦙 ChatGPT can alrea...

Does paying 5x more per page actually get you better document extraction? We ran the data to find ou...

Document extraction just got a new gear. ⚡️ Introducing Turbo mode for Extract, our fastest way yet...

ExtractBench is now live on @Kaggle. It tests schema-guided document extraction on the documents mo...

LlamaParse is officially a verified connector on Claude! 🦙 Claude is perfectly fine for the occasi...

Static embeddings offer unmatched throughput, but they also suffer from accuracy loss compared to tr...

Most extraction tools treat spreadsheets like PDFs. They flatten the file into text or markdown, the...

Our 2nd founder dinner in SF co-hosted by @jerryjliu0 and @GuangyuRobert at @Fundamental - the team ...

Every extraction API demos well on a clean invoice. But what about the scanned form, the nested tabl...

Our 2nd founder dinner in SF co-hosted by @jerryjliu0 and @GuangyuRobert at @Fundamental - the team ...

Our CEO @jerryjliu0 is speaking at @CoreWeave Fully Connected 2026 in SF. 🎤 "Automating Document W...

Most parsers treat tracked changes as noise. Result: a deleted clause comes back as live text, and y...

Most document extraction APIs can't tell you where a value came from. For ExtractBench, we scored g...

Get ready to break racks and agents. If you're in SF, next Thursday we're co-hosting a billiards n...

🦙☀️ It’s a hot Llama summer, and it pays to be a PRO at document parsing. For a limited time, upgra...

Every document extraction system has a perception blind spot. We mapped them. For ExtractBench, we ...

The most dangerous document extraction failure isn't a wrong value. It's a missing row that looks li...

Introducing 𝗘𝘅𝘁𝗿𝗮𝗰𝘁𝗕𝗲𝗻𝗰𝗵: the most comprehensive benchmark for information extraction fr...

SF builders, it's time to ditch your agents and head out on a 🏃 Join us for a run tomorrow with t...