LynseDB 0.2.0

A lightweight, server-optional, multi-end compatible, Rust-powered vector database deployable locally or remotely.

LynseDB is a lightweight vector database with a Python-first API and a Rust storage/search backend. It can run embedded in one Python process or as an HTTP server for multi-process and production deployments.

Features

• Server-optional: local embedded mode and remote HTTP mode share the same high-level Python API.
• Dense vector search with flat, HNSW, IVF, DiskANN, and quantized index families.
• Metadata filtering with SQL-like where expressions.
• Named vector fields for multimodal records.
• Sparse, BM25 text, and hybrid retrieval.
• ResultView return objects with NumPy arrays and dataframe/JSON conversion helpers.
• HTTP health checks, metrics, OpenAPI schema, API key auth, snapshots, export/import, and compaction.

Install

Python 3.9 or newer is required.

Native Linux and macOS environments are supported. Native Windows environments are not supported; on Windows, run LynseDB inside WSL 2 (Windows Subsystem for Linux) or use Docker.

pip install LynseDB

Quick example

import numpy as np
import lynse

client = lynse.VectorDBClient(uri="./lynsedb-data")
db = client.create_database("demo", drop_if_exists=True)
collection = db.require_collection("documents", dim=4, drop_if_exists=True)

with collection.insert_session() as session:
    session.bulk_add_items(
        [
            ([0.10, 0.20, 0.30, 0.40], 1, {"title": "intro", "lang": "en"}),
            ([0.11, 0.19, 0.29, 0.39], 2, {"title": "guide", "lang": "en"}),
            ([0.80, 0.10, 0.20, 0.10], 3, {"title": "notes", "lang": "fr"}),
        ],
        enable_progress_bar=False,
    )

collection.build_index("FLAT-L2")

result = collection.search(
    np.array([0.10, 0.20, 0.30, 0.40], dtype=np.float32),
    k=2,
    where="lang = 'en'",
    return_fields=True,
)

print(result.to_list())

Local and remote modes

Local embedded mode:

client = lynse.VectorDBClient(uri="./data")

Remote HTTP mode:

lynse serve --host 0.0.0.0 --port 7637 --data-dir ./server-data

client = lynse.VectorDBClient("http://127.0.0.1:7637")

With API key auth:

lynse serve --host 0.0.0.0 --port 7637 --data-dir ./server-data --api-key your_key

client = lynse.VectorDBClient("http://127.0.0.1:7637", api_key="your_key")

Use local mode for notebooks, tests, and single-process apps. Use remote mode when multiple processes need to share the same database.

Docker

docker run -p 7637:7637 -v lynsedb-data:/data birchkwok/lynsedb:latest
docker run -p 7637:7637 -e LYNSE_API_KEY=your_key -v lynsedb-data:/data birchkwok/lynsedb:latest

On Windows, use this Docker image or install/run LynseDB from a Linux environment in WSL 2.

Documentation

• Quickstart
• Connect and deploy
• Add vectors
• Search and filter
• Indexing guide
• Named, sparse, and hybrid search
• Backup and maintenance
• Field filters
• ResultView

Stability notes

LynseDB is still evolving. Pin package and server image versions for deployments. For concurrent production access, prefer the HTTP server over sharing one local data directory across independent Python processes.