Chroma

Chroma is an open-source (Apache-2.0) AI-native database for embeddings and retrieval: store documents with metadata, auto-embed with OpenAI/Cohere/Hugging Face/sentence-transformers, and query with dense, sparse, and hybrid vector search, full-text and regex search, and metadata filtering. It also supports multi-modal retrieval over text, images, and audio.

The Python client offers an in-memory mode for prototyping and a persistent mode or client-server deployment for real workloads; there are TypeScript and Rust clients too, plus a managed Chroma Cloud option. Backed by Quiet Capital ($18M seed, 2023, per SiliconANGLE).

1. Install the client: pip install chromadb (verified in the repo README).
2. Start with the in-memory client for prototyping:

```python

import chromadb

client = chromadb.Client()

collection = client.create_collection(name="my_collection")

collection.add(documents=["Paris is the capital of France."], ids=["doc1"])

results = collection.query(query_texts=["What is the capital of France?"], n_results=2)

```

1. For anything persistent, switch to chromadb.PersistentClient(path="/data") or run Chroma in client-server mode — in-memory data is lost when the program exits (verified in the docs).
2. Bring your own embedding function for a specific provider, or let Chroma's default handle it.
3. Disable the default-on anonymous telemetry via the opt-out documented in the Chroma telemetry docs if it matters for your deployment.