LlamaIndex 🦙

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.

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 → http://opendocrouter.ai

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. https://www.youtube.com/watch?v=MLZ1dUOZG74

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 👇 https://www.llamaindex.ai/blog/why-vlms-can-t-read-forms

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. https://cloud.llamaindex.ai/signup Source: https://www.federalreserve.gov/monetarypolicy/files/fomcprojtabl20260916.pdf

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! ⬇️

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!

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: https://www.bea.gov/sites/default/files/2026-09/gdp2q26-3rd.pdf

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!

This is a huge moment for LlamaIndex and an incredible milestone for our research team: introducing OCRia-1, our new document parsing model. OCRia-1 pushes the Pareto frontier on accuracy, speed, and cost, delivering SOTA results across the most complex of tables, charts, and dense formatting. You can try it out live in LlamaParse, and if you're looking to migrate / processing at scale, we’re offering volume discounts and up to $3,000 in credits to make the switch for the next 48 hours*! http://cloud.llamaindex.ai

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! https://opendocrouter.ai

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: https://www.llamaindex.ai/llamaparse