🧠 ChromaDB Flashcards
🔍 What is ChromaDB?
ChromaDB is an open-source vector database optimized for building LLM-based applications with fast, in-memory search.
🧠 What makes ChromaDB unique?
ChromaDB runs entirely in-process and stores metadata + embeddings, enabling fast local development of AI apps.
📦 What data does it store?
ChromaDB stores documents, metadata, and vector embeddings in memory or persistent storage.
⚙️ How do you use ChromaDB?
Install via pip and integrate it with your Python or LangChain-based application using a simple API.
📚 Is it production-ready?
ChromaDB is ideal for prototyping and small-scale production workloads with fast, lightweight embedding search.
🔧 How does ChromaDB persist data?
You can use ChromaDB with a persistent directory to save and reload collections across sessions.
🧪 Is it good for RAG?
Yes! It’s widely used for Retrieval-Augmented Generation in LLM apps like chatbots and assistants.
🔌 Does ChromaDB work with LangChain?
Yes, ChromaDB is one of the default vector stores supported by LangChain out-of-the-box.
💻 Can you use ChromaDB offline?
Absolutely. Since it’s in-process, you can build local apps without any network or cloud dependency.
🧰 What’s the API like?
ChromaDB uses a simple Python API for creating collections, adding documents, querying vectors, and updating metadata.