Hi there, Llama Lovers! 🦙
Welcome to this week's edition of the LlamaIndex newsletter! We have some exciting updates for you, including the introduction of AgentWorkflow, a new multi-agent workflow for generating visually rich presentations, and a fantastic project for building a travel planner agent. Dive into the details below to maximize your use of these innovative features!
🤩 The Highlights:
- AgentWorkflow Release: Big news! We’ve introduced AgentWorkflow, a high-level system for creating multi-agent systems in LlamaIndex. With flexible agent types, built-in state management, and real-time monitoring, it simplifies the process of defining agents. Learn more in our intro blog post, check out the starter tutorial, and explore the full docs.
- Presenter 🎨🖼️: We’re excited to feature a multi-agent workflow that can generate visually rich presentations. This reference repo includes features like generating Mermaid diagrams, scripts with narration through @elevenlabsio, and rendering presentations as HTML or video. Check it out here and get started with LlamaIndex workflows.
- DeepSeek-R1: The new open source model is the talk of the town! Check out how R1 compares to OpenAI's o1 on RAG tasks in this thread
✨ Framework:
- AgentWorkflows: Our new AgentWorkflow system allows you to define agents easily and provides a robust framework for multi-agent systems. Explore the intro blog post and starter tutorial to get started.
- Document Research Agent: Learn how to build a document research agent using LlamaIndex, inspired by @Google Deep Research. This open-source template breaks down tasks into manageable components. Check out the video here.
- Vercel integration: Did you know you can use LlamaIndex to speed your use of Vercel's ai package in TypeScript? Find out how!
✍️ Community:
- Webinar Recap: Catch the recording of our joint webinar with @Redisinc on building AI agents. It covers the ins and outs of AI agents, best practices for designing agentic systems, and popular tools and frameworks. Watch the full video here.
- Knowledge Graphs: Learn how @memgraphdb and LlamaIndex work together to build agentic graph applications. Read the recap or watch the webinar recording here
- Travel Planner Agent: Build a travel planner agent from scratch! This project by @clusteredbytes shows you how to process user requests and find relevant flights, hotels, and places to visit. Discover the project here and refer to the Workflows guide for more details.
