---
id: chromadb-client
version: "1.5.9"
license: unclear
license_treatment: permissive
maintenance: active
---

# chromadb-client — Chroma Client.

License: permissive · Maintenance: active · Downloads: 571.5K/mo

## What it is and what it does

chromadb-client is a lightweight HTTP-only Python client for the Chroma vector database server. It provides a simple interface to connect to a remote Chroma instance and perform operations like creating collections, adding documents with embeddings and metadata, and querying by semantic similarity. The client handles serialization and network communication via httpx, with built-in retry logic through tenacity and structured data validation via pydantic.

This package is designed for scenarios where you want to keep your application separate from the Chroma server infrastructure—useful for distributed setups, containerized deployments, or when you need a thin client without the full Chroma library overhead. It trades the all-in-one convenience of the full chromadb package for a smaller dependency footprint and clearer separation of concerns.

Use it for:

- Connect a web service or microservice to a centralized Chroma server for semantic search across shared document collections.
- Build a retrieval-augmented generation (RAG) pipeline that queries embeddings stored on a remote Chroma instance.
- Integrate document storage and similarity search into a larger distributed system without embedding the full Chroma library.
- Query vector embeddings with metadata filtering in a production environment where the database runs on a separate server.
- Prototype or test Chroma functionality against a live server without installing the full development library.

## Worth the install?

AI-flagged interpretation of the facts above — verify before relying.

HTTP client library for connecting to a Chroma server to store, query, and manage document embeddings and metadata.

Yes, if you need to connect to a Chroma server from a Python application. The package is actively maintained, has no known vulnerabilities, carries a permissive license, and installs with low friction. It is a straightforward choice for client-side vector database operations in distributed or containerized environments.

## Install

pip install chromadb-client
uv add chromadb-client
poetry add chromadb-client

## Installing chromadb-client

Before you install: Low friction install with 14 runtime dependencies, all common data and networking libraries. Package is actively maintained with recent commits and no known vulnerabilities.

License in practice: Apache 2.0 permissive license allows commercial use, modification, and distribution with minimal restrictions—suitable for most projects.

Quickstart:

pip install chromadb-client

import chromadb
client = chromadb.HttpClient(host="localhost", port=8000)
collection = client.create_collection("docs")
collection.add(documents=["text"], ids=["id1"])
results = collection.query(query_texts=["search"], n_results=2)

Requires a running Chroma server instance accessible at the specified host and port.

Verify before relying:

- Whether the client supports all Chroma server API features or a subset.
- Performance characteristics and latency expectations for typical query workloads.
- Backward compatibility guarantees across server and client version mismatches.

## Package facts

- License: not declared (permissive)
- Python support: supports_current
- Install friction: low
- Maintenance: active
- Downloads: 571.5K/month (top 15,000 on PyPI)
- Known vulnerabilities: none known

## Tags

chroma http client, vector database client, embedding storage client, document retrieval with metadata, remote vector store, chroma server connection, semantic search client, vector-database, rag, embeddings

[View on SkillFed](https://skillfed.io/packages/chromadb-client) · [View on PyPI](https://pypi.org/project/chromadb-client/)