Build RAG using Llama 3.1
summary
Step‑by‑step builds a working RAG pipeline that uses Llama 3.1 8B instant via Grok Cloud API key, converts a sample Paul Graham essay into embeddings with HuggingFace, chunks text (1000 token chunks with overlap), stores vectors in a VectorStoreIndex from LlamaIndex, and exposes a query engine that returns precise answers to factual and contextual questions. Demonstrates creating and revoking free API keys, shows the exact code flow, and the example Q&A works live.
AI-written summary · about
chapters (13)
- 0:00Introduction
- 13:28Session overview
- 15:54NLP fundamentals
- 21:14Retrieval vs modern RAG
- 24:52RAG step by step
- 31:38LLM models for RAG
- 38:25Llama 3.1 intro
- 40:54Llama Index for RAG
- 46:56Hands-on demo setup
- 50:51API key setup
- 1:01:10Building RAG pipeline
- 1:06:33Demo results
- 1:07:17Q&A
related talks
42mTake Your RAG To The Max!
01h 00mKeynote: Building and Productionizing RAG
35mAdvanced RAG Techniques with LlamaIndex
42mTo RAG or not to RAG
31mEnhancing RAG-based apps by constructing and leveraging knowledge graphs with open-source LLMs
48mCrafting your own RAG system: Leveraging 30+ LLMs for enhanced performance
01h 34mHow to Construct and Assess Your First LLM Based RAG System
11mBuilding easily deployable RAG applications with Llamastack