LlamaIndex
Get started
Talk to us

© 2024 LlamaIndex

LlamaIndex • Nov 5, 2024

LlamaIndex Newsletter 2024-11-05

Greetings, Llama Fans! 🦙

Welcome to this week’s edition of the LlamaIndex newsletter! We’re bringing you updates on new LlamaParse features, including Continuous Mode for multi-page table stitching and direct Excel output for seamless document processing. This edition also introduces our one-line deployment tool for financial analyst applications with create-llama, and a guide to building multi-agent report generation workflows with integrated human validation. Plus, don’t miss exciting use cases and tutorials from the community.

If you haven't explored LlamaCloud yet, make sure to sign up and get in touch with us to discuss your specific enterprise use case.

🤩 The highlights:

  • New LlamaParse Features: Continuous Mode for seamless multi-page table stitching and direct Excel output for streamlined document parsing. Blogpost, Tweet.
  • Multi-Agent Report Generation Guide: Tutorial on building report workflows with integrated human validation for client/server applications. Notebook, Tweet.
  • One-Line Deployment for Financial Apps: Launch of create-llama’s one-line deployment tool for automating form-filling tasks in financial analysis with LlamaIndex workflows. Tweet.
  • Cohere Multi-Modal Embeddings Integration: Combine images and text in the same vector space to build effective Multi-Modal RAG applications. Docs, Tweet.

🗺️ LlamaCloud And LlamaParse:

  • We've launched new features in LlamaParse: Continuous Mode for seamlessly stitching tables across multiple pages, and Excel spreadsheet output for direct document parsing into spreadsheets, both functioning together for enhanced document processing. Blogpost, Tweet.
  • Guide to Building a Multi-Agent Report Generation Workflow: This tutorial demonstrates how to integrate human validation into every step of generating long-form content using LLMs, featuring a message-passing, streaming architecture suitable for client/server applications. Notebook, Tweet.

✨ Framework:

  • We have launched a one-line deployment tool with create-llama for building full-stack financial analyst applications that automate form filling tasks, such as populating CSVs or Excel files from financial reports. This open-source tool is powered by LlamaIndex workflows and includes a customizable Next.js frontend. Tweet.
  • LlamaIndex's chat-ui: Upgrade your AI projects with popular chat components from create-llama. This React component library features pre-built chat elements, Tailwind CSS customization, and easy integration with Vercel's AI library, including custom widgets, code and LaTeX styling, and a PDF viewer. Tweet.
  • We have integrated Cohere multi-modal embeddings, allowing you to combine images and text in the same vector space, and build effective Multi-Modal RAG applications. Docs, Tweet.
  • We have integrated Open Telemetry with Braintrust Data for enhanced logging and observability in LlamaIndex applications, allowing you to monitor complex, production environments effectively. Docs, Tweet.

💡 Use-cases:

  • Fully local RAG-augmented voice chatbot: A terminal tool by drunkwcodes that lets you upload documents and converse with a context-augmented chatbot with voice outputs, without any external calls. Code, Tweet.
  • Accelerate mechanical maintenance with AI: OilyRAGs, an AI-driven catalog for mechanics, transforms traditional industries with faster parts lookup and a hands-free interface for the marine sector, powered by LlamaIndex, Pinecone, OpenAI, and ArizePhoenix integrations. Blogpost, Tweet.

✍️ Community:

💻 Hackathon:

  • Join us for the 3-day Llama Impact Hackathon in San Francisco from November 8-10, featuring Meta's Llama 3.2 models, a $15,000 prize pool, and a chance to unlock a $100,000 grant; participate in-person or online.
  • Join us for the first LLM x Law Hackathon in Paris on November 17th, co-hosted with CodeX Stanford, Koyeb, LlamaIndex, Neon Database, MistralAI, and GenAI Collective at Station F.

👥 Recruitment: