OmniAI

I can *confidently* say this is one of our best new features! Document extraction is hard. And despite what a lot of companies pitch, LLMs are not 100% accurate. And that’s fine! Businesses work around systems that are less than 100% accurate all the time. The key is knowing when to trust the output and when to bring in a human to double-check. It's about knowing when automate and when to escalate! That’s why we’ve been working hard to roll out confidence scoring on all our document extractions, complete with customizable confidence cutoffs that trigger manual reviews. Why is this a hard problem? Unlike traditional OCR, which mostly deals with recognizing characters and patterns, extraction confidence is a tougher nut to crack. OCR can tell you it’s 99% sure that squiggle is a “B,” but extracting meaning (i.e. like pulling a date, a name, or a contract term from a messy, unstructured document) requires understanding context, intent, and subtle nuances. Layer in handwriting, faded scans, or inconsistent formats, and the challenge gets even bigger. Assigning a reliable confidence score to that process means wrestling with ambiguity in a way OCR never had to. After months of training and fine tuning our confidence model, I’m thrilled to announce this feature is finally live! (And everyone who's asked us for confidence the last 6 months we appreciate the patience ??) It's already been a game changer for our customers, and it's allowing people to build more and more complex automations when they know exactly when to automate and when to escalate. This is live now if in our API dashboard. Just drop a comment if you want an invite to try it out with your documents.

Lets turn a document into an API endpoint! Every PDF would be better as JSON (I think most devs agree on this point), but coming up with a standardized response format is a lot harder than you might expect. Say you're doing financial statement processing, but you're dealing with 50 different document formats. Each one probably has ~90% information overlap, but how do you normalize across different layouts, standards, reporting periods, etc? Before today it was a pretty manual task. Even on our side. We'd flip back and forth between a dozen example docs trying to track down all the edge cases. But realized this was clearly a task an AI would do way better than we could So we revamped out schema builder! Now just pass in the example documents and click the "Suggest Schema" button. And Omni will scan through all the examples and build a normalized template that works across multiple document formats. Took me three minutes to go from a handful of PDFs to a standardized balance sheet API.

one month in our new space in sf, and finally celebrated last friday! ?? P.S. we're hiring founding engineers and 1st growth! our ceo Tyler Maran will grill the best burgers and skewers for you!

Today, we’re officially launching the Omni OCR Benchmark! Are LLMs a total replacement for traditional OCR models? It's been an increasingly hot topic, especially with models like Gemini 2.0 becoming cost competitive with traditional OCR. And we wanted to put some numbers behind it! It's been a huge team effort to collect and manually annotate the real world document data for this evaluation. We've spent more time reading PDFs than I ever thought possible. And we're making that work open source! Our goal with this benchmark is to provide the most comprehensive, open-source evaluation of OCR / document extraction accuracy across both traditional OCR providers and multimodal LLMs. We’ve compared the top providers on 1,000 documents. The three big metrics we measured: - Accuracy (how well can the model extract structured data) - Cost per 1,000 pages - Latency per page A link to the full report + data explorer is below. Check it out!

Today, we’re officially launching the Omni OCR Benchmark! Read the full full benchmark here! https://lnkd.in/eH-fbtqE This benchmark and evaluation data are fully Open Source. - Github: https://lnkd.in/eW9iapmr - Huggingface: https://lnkd.in/einC9Yn7 Our goal with this benchmark is to provide the most comprehensive, open-source evaluation of OCR / document extraction accuracy across both traditional OCR providers and multimodal LLMs. We’ve tested 10 popular providers on 1,000 documents, measuring JSON accuracy, cost per 1,000 pages, and latency per page. Evaluating document parsing is difficult, especially with documents containing charts, handwriting, tables, etc. We hope this benchmark provides an easy way to compare providers for your use case.

We helped an asset manager save 14 hours on every new customer onboarding. Whenever onboarding a new customer, their team used to manually review all of the customer’s financials - their brokerage statement, bank statements, fact sheets, etc. The process took days and was very manual. OmniAI automates this process in minutes! Here’s a demo pulling transactions, balances, and account details from a bank statement. Time is your most valuable asset, don’t spend it copy and pasting.

What do you do when someone sends you a PDF with 12,640 rows you've got to extract? Get ready for some serious copy pasting... This happens WAY more often than you'd think. And clearly this came from a database that someone exported as a PDF. So shouldn't there be a better way? Unfortunately that's just how a ton of enterprise software systems work (EHRs, Inventory systems, ERPs, etc.). And if you need to use that data, you're stuck with whatever format they give you. So companies are spending hundreds of hours of engineering time figuring out how to parse that data. PDF table extraction is pretty hard normally. But there's an entirely different set of problems when you're in the 500+ page range. But good news! This whole process takes about 10 minutes to set up on Omni. And zero engineering required. Just specify what fields you're looking for, and we'll go page by page aggregating data into a single clean table. We can't stop the world from saving databases as PDFs, but at least we can turn it back into useful data!

New year, new space for OmniAI! Today’s our first official day in the new office ?? Super excited to make great memories, close big deals, and host amazing Omni events! We’re still furnishing it, but of course we have all the necessities covered - coffee, monitors,?and whiteboards ?? We’re all about in-person collaboration. Nothing beats building together, even if one person has to be on the giant screen ??

We know LLMs can read a pdf, but are they any good at it? Turns out you need a lot of PDFs to answer that question! Let's talk document extraction. Everyone's building some variant of this (ourselves includes), but how do you know that it actually works? Is the LLM 70% accurate, 90% accurate, better than a human? Turns out it's a pretty tough thing to benchmark. Mostly because you need a LOT of correctly annotated documents, and those aren't easy to find. Especially when it comes to the types of real world document problems that LLMs are supposed to be solving. Traditional OCR benchmarks are looking at text similarity on things like textbooks or receipts. But that falls short in a couple ways. 1. The data sets aren't representative of real world data 2. Those benchmarks are more focused on character recognition than structured extraction. And just having all the letters off of page doesn't get you very far. So lets built a better benchmark! Next week we'll launch our open source VML benchmark. This will include a full validation dataset consisting of: - Document images - Markdown - JSON Schema - Validated JSON response As I mentioned above, our main goal is LLM based document extraction. To benchmark this, we’ll be validating the two most common patterns. ?????????? ? ???????? ? ???????????????????? This is the most common workflow. You run OCR on a document, and pass the resulting text to an LLM along with the JSON schema for extraction. We will be only be evaluating providers on their OCR accuracy, and using GPT-4o structured output for the extraction. ?????????? ? ???????????????????? For multimodal LLM providers, we will run a separate test of direct extraction without the OCR step (i.e. GPT 4o & Anthropic PDF). So far we've added the following providers: - OmniAI (of course!) - Azure Document Intelligence - AWS Textract - Google Document AI - Unstructured - GPT 4o Vision - Claude Sonnet 3.5 - Llama 3.3 Vision - Deepseek R1 Let me know in the comments if you want someone else added to the list:

wow 7 day from my last post and we've gone from 8,000 to 9,000 stars on Zerox! Didn't have time to ship the extra features I promised last week ?? But coming soon! - Structured schema extraction - Edge detection & cropping (improves cell phone pictures of documents) - More model options (including Deepseek and Qwen!) - Dockerized deployment for Zerox