LlamaIndex Newsletter 2025-02-11

Hi there, Llama Enthusiasts! 🦙

Welcome to this week's edition of the LlamaIndex newsletter! We're thrilled to share some exciting updates, tutorials, and community highlights from the past week. Dive in to discover the latest advancements in AI agents, document processing, and more!

🤩 The Highlights:

• Building a YouTube Research Agent: Check out this fantastic tutorial on creating a YouTube research agent from scratch using Gemini Flash 2.0. It integrates tools from @composiohq and @llama_index for enhanced functionality. Tutorial.
• DocumentContextExtractor: Open source works! This technique improves RAG accuracy by providing more context about document chunks before indexing. Now part of @llama_index! Details.
• Comparing Models: Explore @akshay_pachaar’s thread comparing the latest closed vs. open-source models for agent applications, featuring @llama_index and more. Thread.

🗺️ LlamaCloud & LlamaParse:

• Transform PDFs to PowerPoint: Discover how to turn your PDF documents into dynamic PowerPoint presentations using an open-source web app built with @llama_index and LlamaParse. Repo.

✨ Framework:

• Gemini 2.0 Support: Gemini 2.0 from @google is now available, and we have day 0 support! Get the latest integration package to start using it. Announcement.
• LlamaParse Enhancements: LlamaParse now supports Gemini 2.0 Flash, offering high-quality document processing at a fraction of the cost. Notebook.

✍️ Community:

• SciAgents: Check out this automated scientific discovery system using multi-agent workflows and knowledge graphs, leading to innovative research hypotheses. Repo.
• DeepSeek Forum: Join @aicampai for a virtual forum exploring Deepseek and its integration into workflows, featuring presenters from major tech companies. Event Details.
• YouTube Video Summarization Bot: Engineer @KaranVaidya6 built a bot that polls for new YouTube videos, summarizes them, and posts to Slack, email, or Discord. Details.
• Agentic RAG Tutorial: Learn how to build your first RAG application with @llama_index through a comprehensive video tutorial. Watch Here.

Thank you for being part of our community! We look forward to bringing you more updates and innovations in the coming weeks. If you have any questions or suggestions, feel free to reach out!

Best, The LlamaIndex Team