LlamaIndex

Last week was #MSIgnite and we had some big announcements about LlamaParse, LlamaCloud and Microsoft Azure AI! Catch the video of the breakout session from Farzad Sunavala and Laurie Voss with a great demo of LlamaParse and all the things it can do, like: ?? Multimodal parsing ?? Multimodal querying with Azure AI Search ?? Image understanding and query answering The video kicks off with an introduction to RAG and LlamaIndex, so if that's familiar to you skip to the 22-minute mark for the demo! https://lnkd.in/gKarv3aT

Build a quality-aware documentation chatbot using LlamaIndex for RAG and AIMon for monitoring ?? ?? LlamaIndex handles document ingestion and retrieval, using Milvus as the vector store. ?? AIMon continuously monitors LLM outputs, detecting issues like hallucinations and context quality problems. ?? The integration creates a robust chatbot that can identify and fix quality issues in real-time, ensuring reliable responses. https://lnkd.in/gQBJE4GJ

Enhance the performance of your RAG pipeline with fully in-memory operations using CXL memory! CXL allows you to dramatically expand the amount of memory available to your RAG application. This research from MemVerge shows that you can combine CXL memory management and LlamaIndex's unique features like multi-source Simple Composable Memory to significantly boost Queries Per Second with minimal loss of latency. Read the full analysis and benchmark results: https://lnkd.in/gCCvdyyA

We're excited to announce the availability of Microsoft Azure OpenAI endpoints in LlamaParse! ?? LlamaParse is the world's best parser of complex document formats, and now you can use it with your own API endpoints, allowing for enterprise-grade security and compliance for sensitive workloads. Build a cohesive RAG workflow using LlamaCloud, Azure AI Search, and Azure OpenAI with step-by-step guidance and example code: https://lnkd.in/gYmh3kQt

Boost your RAG system's performance before going live by using Ragas to evaluate and optimize! ?? Understand key metrics for RAG evaluation ??? Use LlamaIndex, Ragas, and Literal AI to build and assess your system ?? Analyze context precision, recall, and answer relevancy ?? Discover practical tips to enhance RAG performance Read the full guide here: https://lnkd.in/eBFMiivx

? Last Call! Don’t miss today’s Community Call with Laurie Voss from LlamaIndex and Matea Pe?i? from Memgraph as they dive into the exciting Memgraph x LlamaIndex integration. ?? In this session, you will: ? Get an an?overview of the LlamaIndex framework, focusing on building knowledge graphs from unstructured data ?Explore?advanced retrieval methods?that enable efficient information extraction ?Understand Memgraph’s role in this process - how it?integrates with LlamaIndex?to transform unstructured data into a queryable knowledge graph ?? https://lnkd.in/dM6jmKKs

Learn how Arcee.ai processed millions of pages of NLP research papers using LlamaParse, creating a high-quality dataset for their AI agents: ?? Efficient PDF-to-text conversion, preserving complex elements like tables and equations ?? Flexible prompt system for refining extraction tasks ?? Improved accuracy through iterative prompt adjustments See how LlamaParse outperformed traditional OCR and open-source alternatives in handling intricate scientific content in our case study: https://lnkd.in/gK573Sjt

Join us and MongoDB on December 5th at 9am Pacific for a webinar exploring how to take your RAG applications from basic to agentic! Laurie Voss, VP of Developer Relations at LlamaIndex, and?Anaiya Raisinghani, Developer Advocate at MongoDB will provide you with: ?? Insights into integrating LlamaIndex and MongoDB Atlas ?? Expert tips on building advanced RAG capabilities ?? A deep dive into optimizing relevance with hybrid search Register here: https://lnkd.in/g5GkCfXR

Learn about the state of AI and win a MacBook Pro M4! We've teamed up with Vellum, Fireworks AI, and Weaviate to launch a simple 4-minute survey about the AI tools you use to build your products, covering: ?? Your AI development journey ?? Your team and technology ?? Challenges you've faced ?? Production use-cases ?? AI's impact and your future plans The results will be published in January 2025 and you'll get early access as thanks for your participation! https://lnkd.in/gx88vHaa

create-llama is the easiest way to get started using LlamaIndex. v0.3.15 improves the DX for Python devs: the Next.js frontend is now in a subdir served by the FastAPI backend. Watch below how simple it is to get started. ??? $ npx create-llama ?? https://lnkd.in/gNGyDq5m