LlamaIndex Newsletter 2024-08-06

Greetings, Llama Lovers! 🦙

Welcome to this week’s edition of the LlamaIndex newsletter! We’re excited to share our latest updates including dynamic features like LlamaIndex Workflows and retrieval capabilities in LlamaCloud. Check out our in-depth guides, tutorials, and the upcoming webinars that will help you make the most of these new developments.

🤩 The highlights:

1. LlamaIndex Workflows Launched: LlamaIndex Workflows, a new event-driven architecture for building multi-agent applications, supports batching, async operations, and streaming. Agents subscribe to and emit events for complex, readable, Pythonic orchestration. Blogpost, Tweet.
2. Dynamic Retrieval Feature in LlamaCloud: A new feature in LlamaCloud now supports dynamic retrieval for QA assistants, enabling both chunk-level and file-level document retrieval based on query similarity to intelligently route queries. Blogpost, Notebook, Tweet.
3. LongRAG LlamaPack: LongRAG is now available as a LlamaPack in LlamaIndex, utilizing larger document chunks and long-context LLMs for more effective synthesis. Notebook, Tweet.

✨ Feature Releases and Enhancements:

• We have launched LlamaIndex Workflows, a new event-driven way to build multi-agent applications where each agent acts as a component that subscribes to and emits events, allowing for complex, readable, and Pythonic orchestration with enhanced support for batching, async operations, and streaming. Blogpost, Tweet.
• We have introduced a new feature in LlamaCloud to improve your QA assistant with our latest capability for dynamic retrieval, allowing both chunk-level and file-level retrieval. This feature enables the retrieval of entire documents based on query similarity, which supports building agents that can intelligently route queries based on their content. Blogpost, Notebook, Tweet.
• We have launched LongRAG as a LlamaPack in LlamaIndex. LongRAG simplifies retrieval by using larger document chunks and leveraging long-context LLMs for synthesis. Notebook, Tweet.

🗺️ Guides:

• Guide to building a ReAct agent from scratch using LlamaIndex workflows.
• Guide to Building an Event-Driven RAG Pipeline with LlamaIndex, featuring distinct event-driven steps for retrieval, reranking, and synthesis, enhanced with graph tracing and async processing.
• Guide to MLflow in LlamaIndex to manage, deploy, and monitor your genAI applications with MLflow's tracking, packaging, evaluation, and tracing capabilities.

✍️ Tutorials:

• Pavan Kumar’s tutorial on Building Smarter Agents using LlamaIndex Agents and Qdrant’s Hybrid Search.
• Farzad Sunavala’s tutorial on RAG Observability and Evaluation with Azure AI Search, Azure OpenAI, LlamaIndex, and Arize Phoenix.
• Composio’s tutorial on building a PR review agent using Composio's GitHub/Slack tools and LlamaIndex agent abstractions.
• Benito Martin’s tutorial on Scaling a LlamaIndex and Qdrant Application with Google Kubernetes Engine.
• Chew Loong Nian’s tutorial on Transforming Metadata Extraction for Enhanced RAG Queries using LlamaExtract.
• Pavan Kumar’s tutorial on Practical Implementation of Agentic RAG Workflows with Llama-Index and Qdrant.
• AI21 Labs tutorial on using Jamba-Instruct Model with LlamaIndex.

🎤 Webinars And Hackathons:

• Join us for a webinar on August 8th with Dedy Kredo from CodiumAI on using RAG with LlamaIndex to help build a code generation solution that’s contextually aware of the right elements of source code.
• Join us on RAG Hack Night at GitHub with Weaviate, Neosync, Arize AI on August 13th.