LlamaIndex Newsletter 2024-04-09

Hello, LlamaIndex members! 🦙

Welcome to another thrilling weekly update from LlamaUniverse! We're excited to present a variety of outstanding updates, including Anthropic's Function Calling, Cookbooks, RankLLM, Guides, Tutorials, and much more.

🤩 The highlights:

• Anthropic's Claude Function Calling Agent: Enhance QA/RAG and workflow automation with advanced tool calling in an agent framework. Notebook, Tweet.
• RankLLM Integration: RankLLM is an open-source LLM collection for reranking, surpassing GPT-4 based alternatives is now integrated with LlamaIndex. Notebook, Tweet.
• LlamaIndex + MistralAI Cookbook Series: Launched a cookbook series with MistralAI for building diverse RAG applications, from basic to advanced, with distinctive methods and abstractions. Cookbooks, Tweet

✨ Feature Releases and Enhancements:

1. We have introduced the Anthropic’s Claude Function Calling Agent, leveraging advanced tool calling capabilities within an agent framework for enhanced QA/RAG and workflow automation. Notebook, Tweet.
2. RankLLM (by Ronak Pradeep) integration with LlamaIndex - an open-source LLM collection fine-tuned for reranking, offering top-notch results and outperforming GPT-4 based rerankers. Notebook, Tweet.
3. We have launched the LlamaIndex + MistralAI Cookbook Series for creating a range of RAG applications, from simple setups to advanced agents, featuring unique abstractions and techniques. Cookbooks, Tweet
4. We launched create-llama for building full-stack RAG/agent applications with a single CLI command, akin to create-react-app, for a comprehensive chatbot setup including tool use. Tweet.

🎥 Demos:

• AutoRAG by Marker-Inc-Korea: Streamline RAG pipeline optimization with an automated three-step process, from data preparation to evaluation and optimal pipeline adoption, enhancing the efficiency of the RAG pipeline using LlamaIndex.

🗺️ Guides:

• Guide to Building Advanced RAG with Temporal Filters: Learn how to enhance your RAG pipeline with time-based metadata for more effective financial report analysis using LlamaIndex and KDB.AI vector store.
• Guide to Adaptive RAG for dynamically selecting RAG strategies based on query complexity, enhancing efficiency across varying question types.

✍️ Tutorials:

• (λx.x)eranga’s tutorial on the step-by-step process for building RAG with local models (LlamaIndex, Ollama, HuggingFace Embeddings, ChromaDB) and wrapping it all in a Flask server.
• Ivan Ilin’s video tutorial on iki.ai - an LLM-powered digital library, for organizing, and sharing information within teams or organizations.
• Tutorial on scaling LLM Applications with Koyeb on deploying a full-stack RAG application globally without infrastructure setup, using Koyeb, LlamaIndex.TS, and TogetherAI.
• Ankush Singal's tutorial on Building Multi-Document Agents with LlamaIndex covers advanced multi-document agent concepts, where documents serve as sub-agents enabling complex QA, semantic search, and summarization.
• Rohan’s tutorial on building a Full-Stack RAG application that streams intermediate results to visual UI components with event queues and server-side events.
• Hanane Dupouy's tutorial on building a Finance Agent using an LLM with Yahoo Finance and LlamaIndex abstractions to analyze financial data for publicly traded companies, covering everything from balance sheets to stock recommendations.

🎥 Webinars:

• Webinar with Daniel Huynh ****featuring LaVague - an agent that can navigate the web in your Jupyter/Colab notebook.
• Webinar with Logan Kelly featuring CallSine that utilizes LlamaIndex abstractions and LLMs for personalized sales outreach.