942 stars and a notebook for every LLM framework

What it does This repository is a curated collection of Jupyter notebooks demonstrating how to use Weaviate’s vector database with third-party tools, native features, and cloud services. It covers everything from basic hybrid search and product quantization to multi-tenancy and media search with nearImage and nearVideo. The repo also includes ready-to-use datasets and examples of Weaviate’s newer services like the QueryAgent and TransformationAgent.

The interesting bit The integrations table reads like a census of the modern AI stack, spanning LLM frameworks, cloud hyperscalers, data platforms, and observability vendors. Rather than leaving developers to wrestle with API docs in isolation, it provides working glue code for each pairing.

Key highlights

• Covers integrations with LangChain, LlamaIndex, DSPy, CrewAI, Haystack, and numerous other LLM frameworks
• Demonstrates Weaviate-specific features like ColBERT-style multi-vector embeddings, reranking, and product quantization
• Includes Weaviate Services examples such as QueryAgent and TransformationAgent
• Provides ready-to-use datasets for ingestion testing
• Maintains a contributor guide for converting notebooks into official documentation markdown

Caveats

• The README explicitly notes this is an “ongoing project” with frequent updates, so expect churn and shifting contents
• It is fundamentally a collection of example notebooks and glue code, not a standalone library or framework

Verdict Worth bookmarking if you are building search or generative pipelines with Weaviate and want working patterns before writing your own boilerplate. Skip it if you are looking for a packaged SDK rather than reference implementations.