Learn how to build a serverless RAG application with LlamaIndex and Azure OpenAI! This comprehensive guide by Wassim Chegham walks you through creating an AI-powered app that leverages your own business data for enhanced responses. 🔍 Understand RAG architecture and LlamaIndex implementation 🛠️ Set up a sample application using TypeScript or Python 🚀 Deploy your app to Azure using Azure Developer CLI Step-by-step tutorial covers: • Data ingestion and vectorization • Creating and using vector indexes • Configuring Azure OpenAI access • Implementing user workflows Check out the detailed guide and open-source sample code: https://lnkd.in/gNPZ4JWJ
LlamaIndex
Technology, Information and Internet
San Francisco, California 198,789 followers
The central interface between LLMs and your external data.
About us
The data framework for LLMs Python: Github: https://github.com/jerryjliu/llama_index Docs: https://docs.llamaindex.ai/ Typescript/Javascript: Github: https://github.com/run-llama/LlamaIndexTS Docs: https://ts.llamaindex.ai/ Other: Discord: discord.gg/dGcwcsnxhU LlamaHub: llamahub.ai Twitter: https://twitter.com/llama_index Blog: blog.llamaindex.ai #ai #llms #rag
- Website
-
https://www.llamaindex.ai/
External link for LlamaIndex
- Industry
- Technology, Information and Internet
- Company size
- 2-10 employees
- Headquarters
- San Francisco, California
- Type
- Public Company
Locations
-
Primary
San Francisco, California, US
Employees at LlamaIndex
Updates
-
Need super-fast responses from your LLM? How does 1800 tokens per second for Llama 3.1-8b sound? That's the fastest in the world! Speed is critical to any application, but LLMs can be slow, especially if you have multiple round trips in a complex agentic system. Cerebras Systems has some ground-breaking new tech that also offers 450t/s on Llama 3.1-70b. It's launching today, and we have day 0 support! Check out the Cerebras integration on LlamaHub: https://lnkd.in/gCmr8Wuq And learn more at their site: https://cerebras.ai/
-
Reminder: we're hosting our second RAG-a-thon in cooperation with Pinecone in 6 weeks time with over $7k in cash prizes and counting! Hosted at the 500 Global offices in Palo Alto, the hackathon will run a whole weekend from October 11-13, so you have plenty of time to register and think of ideas! As always we love to see your RAG applications, and this time around we're particularly interested in seeing what you can build with agentic solutions and our new workflows. Current prizes are $7000 in cash and expected to rise as we get more sponsors on board! https://lnkd.in/g5z3pciV
-
We post a lot of content here at LlamaIndex and our newsletter is a great way to keep tabs on it! But the actual process of writing our newsletter, which summarizes everything we tweet in the past week, was a ton of work for us. Laurie Voss figured out how to use the power of LLMs and our TypeScript framework, LlamaIndex.TS, to cut down writing time from hours to minutes. It's a NextJs app hosted on Vercel and should have some tips for getting your projects to work effectively! Check out the open-source repo here: https://lnkd.in/g5836Zbf Or the demo video here: https://lnkd.in/gNTiRVYG
-
Building a knowledge assistant oftentimes requires connecting to lots of sources of data (e.g. vector db + SQL), but a big issue in enterprise LLM dev is data silos ⚠️ - every new data source requires separate authentication management, and knowledge is spread out between teams. This is a nice tutorial by Lisa N. Cao showing you how to build a universal data agent with LlamaIndex + Apache Gravitino. Gravitino provides you a single source of truth to connect both your vector database and SQL tools. Blog: https://lnkd.in/gjvT9_WU To some extent, we’re also building towards that with RAG pipelines in LlamaCloud! If you’re interested sign up here: https://lnkd.in/gi8dxGnt
-
This weekend, we’re providing a definitive set of tutorials on how to build GraphRAG, step-by-step. First, check out this video by Fahd Mirza on implementing the core components of GraphRAG using an in-memory implementation: 1. Extract entities and relationships using LLMs 2. Partition graph into communities and generate community summaries 3. Query all communities and synthesize into a final answer. Then, extend this initial in-memory implementation by storing your property graph in a Neo4j graph database. Big shoutout to Ravi Theja Desetty for the notebook Video: https://lnkd.in/gpYZym5N V1 Notebook: https://lnkd.in/gGb8GJeq V2 notebook (with Neo4j): https://lnkd.in/gzphCWeG Source GraphRAG paper by Edge et al. (+ image credits): https://lnkd.in/dQGZn3kC
-
create-llama keeps getting better, now featuring a structured extraction template!
Creating structured responses in your RAG pipeline is now easy thanks to create-llama from LlamaIndex. Use create-llama v0.1.40 and start with the "Structured Extractor" template featuring a great UX, thanks to Reflex. 🖥️ $ npx create-llama 📄 https://lnkd.in/gNGyDq5m
-
Use workflows to run a multi-strategy RAG pipeline! In this latest video from Laurie Voss he covers: ➡️ The concept of event-driven workflows in LlamaIndex ➡️ How to create a multi-strategy workflow that combines naive RAG, high top-K, and re-ranking approaches ➡️ Implementing query improvement through reflection ➡️ Setting up custom events and steps in your workflow ➡️ Using the collect_events feature to synchronize multiple parallel processes ➡️ How to visualize your workflow using built-in diagram generation tools ➡️ Practical insights on when to use different RAG strategies Check it out here: https://lnkd.in/grGvX3sa Or go straight to the notebook: https://lnkd.in/gBWYGMR7
-
LlamaIndex 0.11 is launched! We’ve added literally hundreds of features and bug fixes since 0.10 back in February, and are continuing our push to make LlamaIndex the production-ready platform you want. Headline features: ⭐️ Workflows replace Query Pipelines! ⭐️ In 0.11, our llama-index-core package is 42% smaller! ⭐️ Full support for Pydantic V2! Check out our blog post on other big features and some breaking changes: https://lnkd.in/dKS3X84w
-
We're excited to show off what LlamaCloud can do!
Chat LlamaIndex is now using LlamaCloud as a backend, which makes it a great full-stack example of using LlamaCloud from LlamaIndex. Create multiple chatbots, each backed by a managed index on LlamaCloud. 🖥️ https://chat.llamaindex.ai 📃 https://lnkd.in/g_xQbx-Y Note: If you have used Chat LlamaIndex before, please clear your browser memory for the app.