ChromaDB Integration

Local, embedded vector storage for development and small deployments.

ChromaDB runs embedded or as a lightweight local service, making it a convenient option for local development or small single-node deployments where running Postgres or Qdrant feels like overkill.

Setup in 4 steps

1Install chromadb and run it embedded or as a local server
2Implement a ChromaStore adapter matching the kit's VectorStore interface
3Use it as your VECTOR_STORE in local .env for fast iteration
4Switch to pgvector or Qdrant via env var when you deploy to production

What's included

No external service required for local development — runs in-process
Fast iteration loop when prototyping RAG ingestion and retrieval logic
Same rag.ingest() / rag.query() interface as the production adapters
Good fit for CLI tools, notebooks, or single-tenant desktop-style deployments
Not recommended as a production replacement for pgvector/Qdrant at scale

Code example

example.py

# app/rag/stores/chroma_store.py
import chromadb

class ChromaStore:
    def __init__(self):
        self.client = chromadb.PersistentClient(path="./.chroma")

    async def upsert(self, collection: str, chunks: list[dict]):
        coll = self.client.get_or_create_collection(collection)
        coll.add(
            ids=[c["id"] for c in chunks],
            embeddings=[c["embedding"] for c in chunks],
            metadatas=[c["metadata"] for c in chunks],
        )

    async def query(self, collection: str, embedding: list[float], top_k: int):
        coll = self.client.get_or_create_collection(collection)
        return coll.query(query_embeddings=[embedding], n_results=top_k)

Full documentation: /docs/rag-pipeline

ChromaDB integration pre-wired and ready.

FastAPI AI Kit ships with ChromaDB configured out of the box. No manual setup required.

Ready to ship your AI backend this weekend?

Join developers who skipped weeks of boilerplate and went straight to building.

Read the docs

No subscriptions · One-time payment · Lifetime updates