LlamaIndex

Our extraction agent in LlamaExtract has gotten a huge upgrade in capabilities ??- it can read a complex, multimodal document and extract into a Pydantic schema with 3+ layers of nesting. ?A sneak peek on what’s possible: I fed in an equity research report which is completely tiled over with text, table scans, and charts, and extracted out a fully structured research summary with the following nested fields: 1. Report metadata - high-level company info like name, ticker, price 2. Company overview - describe business models, segments, geographic breakdown 3. Financial metrics - key ratios like EBITDA + margins/revenue growth over time 4. Investment thesis 5. Valuation 6. Risk factors Huge shoutout to Neeraj Pradhan for all the continued work on this. Sign up to LlamaExtract: https://lnkd.in/g9Wpqn7w Example LlamaIndex notebooks here: https://lnkd.in/gsvKqrNV

Build your own AI agent frontend that can connect to the thousands of MCP servers popping up! With LlamaIndex you can easily build your own multi-agent workflow and connect to any MCP server as a tool. Use our higher-level agent workflow abstractions, or build your own workflow from scratch. Our core wrapper interface is: https://lnkd.in/g96DZS3w We're consolidating examples of building both MCP clients and servers in this repo: https://lnkd.in/gwHSXbXw Big shoutout to Laurie Voss for pulling this together.

MCP week continues! In this post, learn how to use LlamaIndex to prepare documents for inclusion in Claude using a pre-built MCP server for Qdrant , using Angular 's documentation as a data set. ?? Set up Qdrant vector database for efficient document storage and retrieval ?? Process and index AI-friendly Angular documentation ?? Configure an MCP server to connect Qdrant to Claude ??? Implement a complete RAG pipeline for smart documentation retrieval This comprehensive guide walks you through each step, from understanding vector embeddings to testing your first retrieval: https://lnkd.in/gKbWfR8s

Learn how to build AI agent systems with LlamaIndex and SkySQL for reliable text-to-SQL conversion without coding! This webinar explores: ?? AI agents that turn natural language into database actions ?? LlamaIndex 's role in streamlining AI app development ?? SkySQL 's approach to improving SQL generation accuracy Key benefits: ? Simplify interaction with complex database systems ? Create AI database agents for reliable text-to-SQL responses ? Integrate AI agents into applications or leverage them through APIs Join us to see LlamaIndex and SkySQL in action and learn from their no-code agent builder implementation. Register now: https://lnkd.in/gtpr8gzg

It's MCP week apparently! Yesterday we showed you how to use LlamaCloud as an MCP server, today see how to use LlamaIndex as a client to any MCP server! MCP allows you to create agents in LlamaIndex that can use any of hundreds of existing MCP servers as tools, radically expanding the utility of the agents you can build. Getting it working in LlamaIndex is just a few lines of code! Check out our open source MCP demo repo here: https://lnkd.in/g9wa-U4h Install the MCP client integration here: https://lnkd.in/gY4ezNni Learn more about Model Context Protocol: https://lnkd.in/ep3AQhqT

AgentQL is a new LlamaIndex integration that brings real-time web data to AI agents and workflows ???? Here's what you can do now: ?? Extract structured data from any web page ?? Retrieve real-time financial and business intelligence ?? Automate web data flows into your workflows without coding AI agents can now see and interact with the open web, not just pre-built APIs. This is game-changing for Claude and smart IDEs like Cursor and WINDSURF. Learn more about AgentQL and their LlamaIndex integration: https://lnkd.in/gd4NSzQ4

RAG 2.0 is really about grounding general purpose agents in proprietary enterprise context. Instead of simply one-shot answering a simple question, the agent can decide how to interleave retrieval with other tools to take way more actions + generate way more sophisticated outputs (reports, slide decks, code, entire full-stack applications) This is a fantastic repo by Laurie Voss showing you how to setup LlamaCloud to index a large knowledge base of your most complex PDFs, and feed it into Claude. Check it out: https://lnkd.in/gC5NnD-a YouTube: https://lnkd.in/gx87kBTv

LlamaCloud can be used as an MCP server! This allows you to bring up-to-the-second data into your workflow as a tool used by any MCP client. In this video we demonstrate how to use an existing LlamaCloud index as a data source for an MCP server used by Claude Desktop, which can then answer questions. The code to do this is open source, and includes full setup instructions! Check it out here: https://lnkd.in/g2F6b7N6

Using document retrieval for a RAG chatbot is boring. Instead generate a full-stack application from your docs, by plugging in document retrieval as a tool into Cursor or Windsurf (Codeium) through MCP. This is a dope example by Marcus Schiesser ?? * We use LlamaCloud to get high-quality parsing/indexing/retrieval over the most complex financial reports. * By plugging it into a code assistant, it can dynamically fetch this context to build an entire interactive dashboard that can do direct comparisons between two different companies. Check out the thread below ?? You can setup any LlamaCloud index through the UI: https://lnkd.in/g9Wpqn7w Here's a repo of examples to get started! https://lnkd.in/gbAYfzHC

?? Want Google Deep Research for your own documents? Now you can! create-llama from LlamaIndex lets you build a deep researcher agent with a single command. Watch how easy it is ?? ??? $ npx create-llama