Pinned Tweet
👏We just released OCRflux — a multimodel PDF-to-Markdown toolkit that handles multi-column layouts, table parsing & cross-page merging.
🔵Built on top of finetuning the multimodal LLM Qwen2.5-VL-3B-Instruct, it delivers 3× higher throughput on a GTX 3090 GPU than the olmOCR-7B baseline model.
🔵It achieves 0.967 Edit Distance Similarity vs olmOCR’s 0.872 vs Nanonets's 0.858.
Try demo:2ly.link/28rzz
GitHub:github.com/chatdoc-com/OCRFl…
Hugging Face:huggingface.co/ChatDOC/OCRFl…
😃Key features in the threads
#OCRFlux #OCR #PDF2Markdown #Qwen #olmocr #OpenSource #Benchmark #DocumentAI
ChatDOC retweeted
👋🏼ADIÓS A CAMBIAR CONSTANTEMENTE DE HERRAMIENTAS PARA PARSEAR PDFs!
@docdotai es el mejor PDFParser ahora mismo!
Es completamente GRATIS.
Te permite utilizar herramientas OCR como:
- glmocr
- liteparse
- mineru
- nanodoc
- paddleocr
Todo en UNO!
- Generación de .md a través de herramientas como nanodoc o paddleocr
- Comparación web. Interfaz visual para ver los resultados y compararlos entre parsers.
- También puedes usarlo con tu agente preferido. En mi caso, en Claude. Puedes usar la SKILL y hacerle preguntas en lenguaje natural.
- Instalación. Super facil. En menos de 2 minutos.
Te dejo la herramienta abajo👇🏼
We’ve officially upgraded ChatDOC'S default model to GPT-5 mini—bringing you faster reasoning and deeper document insights at zero extra cost.
🖼️Plus, we’ve leveled up our Multimodal Question Packages to GPT-5.3 for unmatched accuracy with charts and tables.
💎 Smarter
💎 Faster
💎 No extra costs
Go to ChatDOC.com and feel the difference of #GPT5
#ChatPDF #TechUpdate #ChatDOC #Productivity
📚✨ NEW UPDATE ALERT! ✨📚
Now you can chat with scanned PDFs for FREE!
Engage with photographed study materials like old textbooks, complex forms, scattered notes, or screenshots
Easily extract any information you need with a fully searchable and interactive format.
Simplify your learning by centralizing all resources in one place for cross-file Q&A, data extraction, or even translation.
Try it now: 2ly.link/27e87
Glad ChatDOC Studio is helping 🙌 really made our day!
🚀 ChatPaper Update
We’ve made two exciting upgrades!
1️⃣ Added daily sync of arXiv – Information Retrieval papers 📚
2️⃣ Introduced AI Summary – get a full-text understanding and key takeaways in seconds 🤖
Explore the latest IR research with instant summaries 👉 chatpaper.com/
🚀 Launch your AI Help Desk in minutes with ChatDOC Studio !
Turn your business docs into a 24/7 AI assistant — with answer tracing, brand customization, no-code deployment, and API integration.
💡 Train with your files or website
🔍 Smarter search via Context Retrieval API
📄 Parse PDFs into structured, LLM-ready data
👥 Collaborate in one shared Knowledge Hub
👉 Try it now: chatdoc.studio/
#AI #Chatbot #SaaS #NoCode #Agents #ChatDOC #GPT
We just launched ChatDOC Studio! 🚀 Transform any documents into an expert AI assistant, get highly accurate, traceable answers for every query. Start building today with our FREE plan!
- 1,000 Credits per month
- Create 3 AI apps
- Add 3 Team members
- 100MB of knowledge base storage
linkly.link/2FEon
#AI #Chatbot #NoCode #CustomerSupport #AIAgent #aiforbusinessleaders
📷 Immersive Q&A Mode is here!
Hide the document panel and focus only on the answers you need.
A cleaner, distraction-free way to browse and interact with your documents.
👉Try it now → chatdoc.com/
ChatDOC retweeted
卧槽, 真的 SOTA 了, 正在准备做一个 PDF 转 Markdown 的领域大模型评测, 结果我发现 ChatDOC 的 OCRFlux-3B 在我测试的所有场景均达到了现阶段最理想的效果. 下面直接为大家带来实测:
场景A:
让我们直接来现阶段的 PDF 转化试金石——发票转 Markdown
可以看到, 重要的 购买方, 单价, 价税合计, 销售方均十分准确, 并且没有把印章错误的当成内容. 而且形成的 Markdown 表格行列准确, 甚至还有合并单元格的表格.
以往的测试别说表格, 连内容都不一定识别准确. 更别提生成正确的 Markdown 表格了. 当然也不是没有瑕疵. 如果 OCRFlux 可以把表头和页脚也能准确识别进去就完美了. 现在生成的内容用来识别普票还是专票还是有困难的.
#OCRFlux #ChatDOC #Markdown #PDF
🤩Seriously made our day seeing you dig our multi-lingual &column doc parsing!
👇Everyone come give it a spin and tell us what you think! 2ly.link/27e87
#PDFparsing #OCRFlux #ocr #opensourceai
PDF parsing is still painful because LLMs reorder text in complex layouts, break tables across pages, and fail on graphs or images.
💡Testing the new open-source OCRFlux model, and here the results are really good for a change.
So OCRFlux is a multimodal, LLM based toolkit for converting PDFs and images into clean, readable, plain Markdown text.
Because the underlying VLM is only 3B param, it runs even on a 3090 GPU. The model is available on @huggingface .
The engine that powers the OCRFlux, teaches the model to rebuild every page and then stitch fragments across pages into one clean Markdown file.
It bundles one vision language model with 3B parameters that was fine-tuned from Qwen 2.5-VL-3B-Instruct for both page parsing and cross-page merging.
OCRFlux reads raw page images and, guided by task prompts, outputs Markdown for each page and merges split elements across pages.
The evaluation shows Edit Distance Similarity (EDS) 0.967 and cross‑page table Tree Edit Distance 0.950, so the parser is both accurate and layout aware.
How it works while parsing each page
- Convert into text with a natural reading order, even in the presence of multi-column layouts, figures, and insets
- Support for complicated tables and equations
- Automatically removes headers and footers
Cross-page table/paragraph merging
- Cross-page table merging
- Cross-page paragraph merging
A compact vision‑language models can beat bigger models once cross‑page context is added.
🧵 1/n Read on 👇
Thanks for the shoutout! 🙌 Even the messiest multi-page tables get stitched together perfectly by our magic. Try it yourself at 2ly.link/27e87
PDFの常識を変える?!🤯✨ 新ツール「OCRFlux」が登場しましたよ!
✎. FYIG: xitter.cf/aigclink/status/193818…
PDFをMarkdownに変換する新しいツール「OCRFlux」が発表されたそうです!これがもう、すごい機能満載なんですって!
一番驚いたのは、PDFでページをまたぐ表や段落を、自動で繋げてMarkdownにしてくれること!😲 これって、今まで手作業で調整していた部分ですよね。論文や長い報告書を扱う方には、革命的な機能なんじゃないでしょうか?!
なんと98.3%という驚異の正確率なんですって!✨ しかも、多列レイアウトや、中国語と英語が混ざったドキュメントまで、ちゃんと自然な読み順で変換してくれるそうですよ。グローバルな資料作成にも心強いですね!
他の有名なOCRツールと比べても、ベンチマークで圧倒的な性能を示しているとのこと!そして、GTX 3090などのGPUがあれば、3Bパラメータモデルでめちゃくちゃ高速に処理できちゃうそうです!🚀 作業効率がグンと上がりそうでワクワクしますね!
実際の変換の様子は動画で見られますが、まるで魔法みたいにサクサク動いていましたよ!これがあれば、PDFからのデータ抽出や再利用が格段にラクになりますよね。
例えば、古いPDF資料をMarkdown化して検索しやすくしたり、議事録をPDFでもらってもすぐに編集可能なテキストに変換したり…アイデア次第で色々な使い方ができそうです!
詳しい情報や、もしかしたら試せるデモもあるかもしれませんね!ぜひチェックしてみてください!
PDFを扱う皆さんの救世主になるかもしれませんね!これからが本当に楽しみです!😊
#OCRFlux #PDF変換 #AIツール
ChatDOC retweeted
This PDF-to-Markdown toolkit can convert PDFs and images into clean Markdown, parse complex tables, and much more...
Open this 🧵 ↓ Bookmark for later 🔖
Horizontally split tables across pages? No problem for OCRflux! It recognizes and merges them seamlessly into one page.
Check out👇
Try demo: 2ly.link/27e87
#OCR #pdfparser #pdftomarkdown #OCRFlux #opensourceai
😩Extract tables from PDFs = pure pain? Not anymore! OCRFlux transforms cross-page tables into structured ones, perfectly preserving layouts and auto-removing duplicate headers.
Check out the comparison between olmOCR and OCRFlux 👇
Or try it at: 2ly.link/28rzz
#OCRFlux #ocr #OpenSource #PDFparser #olmOCR
OCRFlux doesn’t just extract text—it understands semantic structure.
👉 Section titles
👉 Multi-Column Layouts
👉 Footnotes
It sees your PDF like a researcher would.
Try demo: 2ly.link/28rzz
#OCR #OpenSource #OCRFlux #olmOCR
🥳OCRFlux shines in accuracy! It achieves significantly higher Edit Distance Similarity (EDS) and Tree Edit Distance-based Similarity (TEDS) scores compared to #olmOCR and #Nanonets.
Try our demo: 2ly.link/28rkt #OCRFlux #OCR #Benchmark






















