Using Weaviate for building Generative AI apps
summary
Open source vector database Weaviate stores embeddings from Cohere/Hugging Face/OpenAI, demo loads 100,000 chunked Wikipedia paragraphs into a WCS instance (free tier 100k objects, 14 days) and runs semantic search (dot‑product / HNSW) plus generative search by passing retrieved docs to ChatGPT either per‑document or as a grouped prompt. Supports multilingual models, batch uploads, Spark streaming, and claims linear scaling up to 1 billion objects in benchmark tests.
AI-written summary · about
chapters (12)
- 0:00Intro
- 0:33Recap
- 4:03Demo
- 8:17Semantic Search
- 14:33Search Results
- 14:48Prompt Engineering
- 15:33Multitenancy
- 17:58Group Task
- 18:48Streaming Indices
- 20:43Vector Search
- 21:08Questions
- 23:08Chatbots
related talks
24mIntroduction to Vector Search with Weaviate
31mBuild your own search application with Weaviate
45mA gentle introduction to Vector Databases
38mAI-powered Vector Search Engine Weaviate
13mSearch through your data with Weaviate
50mWhat Comes After ChatGPT? Vector Databases - the Simple & Powerful Future of ML?
31mVue, You and AI Search
01h 03mWhat comes after ChatGPT? Vector Databases - the Simple and powerful future of ML?