Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

🚀 New in Heptabase: PDF Parser - Extract text, tables, equations & images (even from scanned PDFs) → save as clean Markdown - Ask AI with precise context (page ranges & paragraphs) → responses include link refs back to the source - MAX mode ensures AI reads every word... show more

Heptabase

18,510 subscribers

36,690 views • 1 year ago •via X (Twitter)

Science & Technology Education Arts

Anya Rossi• Live Now

Private livecam show

6 Comments

RR1 year ago

yeah, finally a tool that understands PDFs are just as messy as my college notes

Dustin1 year ago

This new PDF parser feature sounds incredibly powerful for researchers and students alike. The ability to extract and organize content with precision will surely enhance information management and study workflows.

Robinounet1 year ago

Literally me in front of the video: WOWOWOWOW, this is fucking insane.

rey1 year ago

can somebody tell me how it works technically, i'm really curious how they do this.

Ivan Selivanov10 months ago

It doesnt work for Russian texts 😭😭😭 For example, Цель Руководства BABOK reads as LeJb PyKOBOnCTBa BABOK

Saïd Aitmbarek1 year ago

based extraction/content structuration features & great demo! let's launch you guys on whenever it helps

Related Videos

We’ve launched Web Tab in Heptabase! You can now search Google, YouTube, or open any website directly inside Heptabase — take notes side by side, see AI-suggested related cards, or even chat with the page you’re browsing.

We’ve launched Web Tab in Heptabase! You can now search Google, YouTube, or open any website directly inside Heptabase — take notes side by side, see AI-suggested related cards, or even chat with the page you’re browsing.

Heptabase

15,902 views • 11 months ago

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 Hugging Face . 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 👇

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 Hugging Face . 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 👇

Rohan Paul

149,423 views • 1 year ago

Today, we're excited to release the Heptabase CLI — a new way for external AI agents to interact with your local Heptabase knowledge base from the terminal. Once enabled (Settings > AI Features > CLI), any AI coding agent (e.g., Claude Code, Codex) can: - Create, read, and edit note cards and journals - Search your Card Library - Manage tags - Read your AI Tutor courses, lessons, and chat history This is just the first step; we'll keep expanding the CLI surface so AI agents can act on your knowledge at the same depth that you can. If you build something interesting with it, feel free to share with our community!

Today, we're excited to release the Heptabase CLI — a new way for external AI agents to interact with your local Heptabase knowledge base from the terminal. Once enabled (Settings > AI Features > CLI), any AI coding agent (e.g., Claude Code, Codex) can: - Create, read, and edit note cards and journals - Search your Card Library - Manage tags - Read your AI Tutor courses, lessons, and chat history This is just the first step; we'll keep expanding the CLI surface so AI agents can act on your knowledge at the same depth that you can. If you build something interesting with it, feel free to share with our community!

Heptabase

14,894 views • 5 months ago

In the latest version, you can add whiteboards as contexts to AI chat! In AI's response, we’ve introduced reference links to the cards and blocks that the response is grounded on, so you’re able to locate the source content.

In the latest version, you can add whiteboards as contexts to AI chat! In AI's response, we’ve introduced reference links to the cards and blocks that the response is grounded on, so you’re able to locate the source content.

Heptabase

14,816 views • 1 year ago

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 !

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 !

LlamaIndex 🦙

13,467 views • 1 month ago

One of the hardest problems with using AI agents to automate meaningful document work (contracts, KYC, diligence, claims, and more) is not actually building the agent, but building the UI/UX audit trail so the human can understand decisions linking back to the source documents - in PDFs, Powerpoint, Word, etc. This requires really good layout detection and segmentation, that you need to surface as *metadata context* to the AI agent, beyond any document -> markdown conversion. We've built really-nice VLM-capabilities to identify and segment all the elements within your document: from tables to charts to form boxes. You can use this metadata to directly provide experiences where humans can analyze the source document even as decisions are being made. Come check out LlamaParse!

One of the hardest problems with using AI agents to automate meaningful document work (contracts, KYC, diligence, claims, and more) is not actually building the agent, but building the UI/UX audit trail so the human can understand decisions linking back to the source documents - in PDFs, Powerpoint, Word, etc. This requires really good layout detection and segmentation, that you need to surface as metadata context to the AI agent, beyond any document -> markdown conversion. We've built really-nice VLM-capabilities to identify and segment all the elements within your document: from tables to charts to form boxes. You can use this metadata to directly provide experiences where humans can analyze the source document even as decisions are being made. Come check out LlamaParse!

Jerry Liu

19,319 views • 6 months ago

We have just released the first version of our web clipper! It allows you to save and tag web content in Heptabase, and further break it down into atomic knowledge cards.

We have just released the first version of our web clipper! It allows you to save and tag web content in Heptabase, and further break it down into atomic knowledge cards.

Heptabase

27,422 views • 2 years ago

Everyone uploading PDFs to Claude is burning 70,000 tokens before asking a single question. A 20-page document costs between 1,500 to 3,000 tokens per page. Formatting. Broken tables. Images. Junk metadata. Claude processes all of it before you even type “summarize this.” There’s a free fix. It’s called Markitdown. Microsoft built it. 10,000+ GitHub stars. Almost nobody in the Claude community is talking about it. It converts any file PDFs, Word docs, Excel sheets, PowerPoints, YouTube videos into clean Markdown text. Token usage drops by up to 70%. Answers get better, not worse. Claude was trained on millions of Markdown documents. It reads the format natively. You’re not compressing quality you’re removing noise. There’s an MCP server too. Connect it to Claude Desktop once. Every file you upload auto-converts to .md before Claude touches it. You do nothing. It just works. Old way: upload PDF → Claude burns 70k tokens on junk → mediocre answer New way: Markitdown → clean .md → 70% fewer tokens → sharper answer The tool is free. The MCP setup takes 10 minutes. The token savings are permanent. You’ve been paying for Claude to read garbage. Stop.

Everyone uploading PDFs to Claude is burning 70,000 tokens before asking a single question. A 20-page document costs between 1,500 to 3,000 tokens per page. Formatting. Broken tables. Images. Junk metadata. Claude processes all of it before you even type “summarize this.” There’s a free fix. It’s called Markitdown. Microsoft built it. 10,000+ GitHub stars. Almost nobody in the Claude community is talking about it. It converts any file PDFs, Word docs, Excel sheets, PowerPoints, YouTube videos into clean Markdown text. Token usage drops by up to 70%. Answers get better, not worse. Claude was trained on millions of Markdown documents. It reads the format natively. You’re not compressing quality you’re removing noise. There’s an MCP server too. Connect it to Claude Desktop once. Every file you upload auto-converts to .md before Claude touches it. You do nothing. It just works. Old way: upload PDF → Claude burns 70k tokens on junk → mediocre answer New way: Markitdown → clean .md → 70% fewer tokens → sharper answer The tool is free. The MCP setup takes 10 minutes. The token savings are permanent. You’ve been paying for Claude to read garbage. Stop.

Veltrx

109,782 views • 3 months ago

LiteParse Samples 📄🧑‍🏫 LiteParse is our free and fast document parser that can parse text with bounding boxes from any document. I’ve created a new repository that contains some demos on how you can make use of its outputs! ✅ Comparison against PyPDF and PyMuPDF ✅ Visual Citations: search for any keyword and see joined bounding boxes pop up in the source images It should be a default tool to help AI agents OCR any document type (supports PDF, Word, Powerpoint + dozens of other formats), before heavier-weight VLM-based document parsing solutions (like our own LlamaParse). LiteParse Samples repo: LiteParse:

LiteParse Samples 📄🧑‍🏫 LiteParse is our free and fast document parser that can parse text with bounding boxes from any document. I’ve created a new repository that contains some demos on how you can make use of its outputs! ✅ Comparison against PyPDF and PyMuPDF ✅ Visual Citations: search for any keyword and see joined bounding boxes pop up in the source images It should be a default tool to help AI agents OCR any document type (supports PDF, Word, Powerpoint + dozens of other formats), before heavier-weight VLM-based document parsing solutions (like our own LlamaParse). LiteParse Samples repo: LiteParse:

Jerry Liu

13,196 views • 6 months ago

Got the vibe but not the words? Starting today, you can simply show or tell AI Mode about what you're looking for and get rich, visual results to explore. ✨ This experience works by bringing together Search’s world-class visual understanding from Lens and Image search with Gemini 2.5’s advanced multimodal capabilities. When you show AI Mode what you’re searching for it uses our new “visual search fan-out” technique to get a deeper understanding of what’s in an image (including subtle details and secondary objects), and runs multiple queries in the background. This helps it understand the visual context and nuance of your request to deliver relevant visual responses. You can also tell AI Mode what vibe you’re going for (no image needed). For example, ask AI Mode for bedroom design inspo and go back and forth to refine your vision. We’re rolling out this experience in English in the U.S. starting this week. Try it out and let us know what you think!

Got the vibe but not the words? Starting today, you can simply show or tell AI Mode about what you're looking for and get rich, visual results to explore. ✨ This experience works by bringing together Search’s world-class visual understanding from Lens and Image search with Gemini 2.5’s advanced multimodal capabilities. When you show AI Mode what you’re searching for it uses our new “visual search fan-out” technique to get a deeper understanding of what’s in an image (including subtle details and secondary objects), and runs multiple queries in the background. This helps it understand the visual context and nuance of your request to deliver relevant visual responses. You can also tell AI Mode what vibe you’re going for (no image needed). For example, ask AI Mode for bedroom design inspo and go back and forth to refine your vision. We’re rolling out this experience in English in the U.S. starting this week. Try it out and let us know what you think!

Google AI

68,285 views • 1 year ago

In the latest version, you can use the Insight Generator, an early alpha feature that helps you grasp the knowledge structure of a long article ten times faster and quickly navigate to different parts of the knowledge source that are useful or interesting to you.

In the latest version, you can use the Insight Generator, an early alpha feature that helps you grasp the knowledge structure of a long article ten times faster and quickly navigate to different parts of the knowledge source that are useful or interesting to you.

Heptabase

15,347 views • 2 years ago

CLAUDE IS ADDING WATERMARKS In compliance with the new EU AI Act, Claude is adding watermarks to its text outputs. For text outputs, it works as described in this clip. The clip is actually from 2024, from my video on SynthID, Google’s AI watermarking technology that works the same way as Claude’s watermarking tech. Basically, if you remember from our “How does AI work?” video, AI is a word generator, and samples from its generated probability distributions of the next word. With the watermark, the probability distribution of the output text is shifted slightly away from what it would have been without the watermark. Note that the watermark is invisible - you can’t see it with the human eye, since it’s literally just text written slightly differently - and currently, the mechanism to detect this watermark only exists within Anthropic (the team behind Claude) itself. Their published article says they are "working to enable users and other third parties to detect Claude's embedded watermarks.” For images, Claude is adding a signature to the metadata - which is easy to remove. What do you think? Google has been adding watermarks (with SynthID) for a while. Watermarking will now be rolled out industry-wide in compliance with the EU AI Act. SynthID was open source and I had some critiques about it… see below for the 5 minute video.

CLAUDE IS ADDING WATERMARKS In compliance with the new EU AI Act, Claude is adding watermarks to its text outputs. For text outputs, it works as described in this clip. The clip is actually from 2024, from my video on SynthID, Google’s AI watermarking technology that works the same way as Claude’s watermarking tech. Basically, if you remember from our “How does AI work?” video, AI is a word generator, and samples from its generated probability distributions of the next word. With the watermark, the probability distribution of the output text is shifted slightly away from what it would have been without the watermark. Note that the watermark is invisible - you can’t see it with the human eye, since it’s literally just text written slightly differently - and currently, the mechanism to detect this watermark only exists within Anthropic (the team behind Claude) itself. Their published article says they are "working to enable users and other third parties to detect Claude's embedded watermarks.” For images, Claude is adding a signature to the metadata - which is easy to remove. What do you think? Google has been adding watermarks (with SynthID) for a while. Watermarking will now be rolled out industry-wide in compliance with the EU AI Act. SynthID was open source and I had some critiques about it… see below for the 5 minute video.

Harper Carroll

13,218 views • 1 month ago

In the latest version, you can now easily create and insert editor templates!

In the latest version, you can now easily create and insert editor templates!

Heptabase

21,026 views • 2 years ago

You can now search and add multiple cards and web elements from the search bar of the right sidebar!

You can now search and add multiple cards and web elements from the search bar of the right sidebar!

Heptabase

12,580 views • 2 years ago

Introducing LiteParse - the best model-free document parsing tool for AI agents 💫 ✅ It’s completely open-source and free. ✅ No GPU required, will process ~500 pages in 2 seconds on commodity hardware ✅ More accurate than PyPDF, PyMuPDF, Markdown. Also way more readable - see below for how we parse tables!! ✅ Supports 50+ file formats, from PDFs to Office docs to images ✅ Is designed to plug and play with Claude Code, OpenClaw, and any other AI agent with a one-line skills install. Supports native screenshotting capabilities. We spent years building up LlamaParse by orchestrating state-of-the-art VLMs over the most complex documents. Along the way we realized that you could get quite far on most docs through fast and cheap text parsing. Take a look at the video below. For really complex tables within PDFs, we output them in a spatial grid that’s both AI and human-interpretable. Any other free/light parser light PyPDF will destroy the representation of this table and output a sequential list. This is not a replacement for a VLM-based OCR tool (it requires 0 GPUs and doesn’t use models), but it is shocking how good it is to parse most documents. Huge shoutout to Logan Markewich and Clelia Bertelli (🦀/acc) for all the work here. Come check it out: Repo:

Introducing LiteParse - the best model-free document parsing tool for AI agents 💫 ✅ It’s completely open-source and free. ✅ No GPU required, will process ~500 pages in 2 seconds on commodity hardware ✅ More accurate than PyPDF, PyMuPDF, Markdown. Also way more readable - see below for how we parse tables!! ✅ Supports 50+ file formats, from PDFs to Office docs to images ✅ Is designed to plug and play with Claude Code, OpenClaw, and any other AI agent with a one-line skills install. Supports native screenshotting capabilities. We spent years building up LlamaParse by orchestrating state-of-the-art VLMs over the most complex documents. Along the way we realized that you could get quite far on most docs through fast and cheap text parsing. Take a look at the video below. For really complex tables within PDFs, we output them in a spatial grid that’s both AI and human-interpretable. Any other free/light parser light PyPDF will destroy the representation of this table and output a sequential list. This is not a replacement for a VLM-based OCR tool (it requires 0 GPUs and doesn’t use models), but it is shocking how good it is to parse most documents. Huge shoutout to Logan Markewich and Clelia Bertelli (🦀/acc) for all the work here. Come check it out: Repo:

Jerry Liu

257,589 views • 6 months ago

MICROSOFT QUIETLY SHIPPED THE RAG PIPELINE EVERYONE KEEPS TRYING TO BUILD FROM SCRATCH it is called Kernel Memory, and it turns any pile of files into an AI memory that answers with citations. drop your data in and Microsoft's pipeline does the rest: > ingest PDFs, Word, PPT, Excel, web pages, even images > extract the text, chunk it, embed it into vectors > store the vectors with metadata and keep the originals in blob storage > ask a question and it pulls the right chunks with semantic plus hybrid search then it hands the answer back grounded, with links to the exact source and full token usage. and it plugs into whatever you already run: > Azure OpenAI, OpenAI, Ollama, or Anthropic for the models > Qdrant, pgvector, Redis, or Azure AI Search for the vectors > every step is a swappable handler, so you can replace any part with your own one heads up: Microsoft archived the repo in June 2026 and ships it as a reference blueprint, not production. but as a map for how memory and RAG are actually built, nothing is cleaner. run it in one command: docker run kernelmemory/service bookmark this. Microsoft already drew the blueprint, you just build on it.

MICROSOFT QUIETLY SHIPPED THE RAG PIPELINE EVERYONE KEEPS TRYING TO BUILD FROM SCRATCH it is called Kernel Memory, and it turns any pile of files into an AI memory that answers with citations. drop your data in and Microsoft's pipeline does the rest: > ingest PDFs, Word, PPT, Excel, web pages, even images > extract the text, chunk it, embed it into vectors > store the vectors with metadata and keep the originals in blob storage > ask a question and it pulls the right chunks with semantic plus hybrid search then it hands the answer back grounded, with links to the exact source and full token usage. and it plugs into whatever you already run: > Azure OpenAI, OpenAI, Ollama, or Anthropic for the models > Qdrant, pgvector, Redis, or Azure AI Search for the vectors > every step is a swappable handler, so you can replace any part with your own one heads up: Microsoft archived the repo in June 2026 and ships it as a reference blueprint, not production. but as a map for how memory and RAG are actually built, nothing is cleaner. run it in one command: docker run kernelmemory/service bookmark this. Microsoft already drew the blueprint, you just build on it.

NO1ennn

85,393 views • 1 month ago

Launching full support for Even Realities G2 smart glasses in MentraOS. With Mentra, you can run multiple smart glasses apps at once. You: in a meeting in Tokyo on 82nd floor. You're running translation on your smart glasses and understanding every word. Your Notetaker is running in the background capturing ideas as they're spoken. Your proactive AI agent is listening in, hears someone ask a question, and automatically overlays the answer on your vision. That's what's Mentra unlocks. MentraOS is the open source OS for smart glasses. We give users an app store and allow developers to write one app that runs on any pair of smart glasses. It's 100% open source, private, and free. How to use: Download the Mentra app, connect your G2, and get instant access to captions, AI notetaker, Merge proactive AI, translation, and other apps. You can switch back to the Even app at anytime. If you want to unlock your Even G2, then downalod Mentra on the Google or Apple App Store.

Launching full support for Even Realities G2 smart glasses in MentraOS. With Mentra, you can run multiple smart glasses apps at once. You: in a meeting in Tokyo on 82nd floor. You're running translation on your smart glasses and understanding every word. Your Notetaker is running in the background capturing ideas as they're spoken. Your proactive AI agent is listening in, hears someone ask a question, and automatically overlays the answer on your vision. That's what's Mentra unlocks. MentraOS is the open source OS for smart glasses. We give users an app store and allow developers to write one app that runs on any pair of smart glasses. It's 100% open source, private, and free. How to use: Download the Mentra app, connect your G2, and get instant access to captions, AI notetaker, Merge proactive AI, translation, and other apps. You can switch back to the Even app at anytime. If you want to unlock your Even G2, then downalod Mentra on the Google or Apple App Store.

cayden 凯登

39,407 views • 4 months ago