---
id: tiledb
version: "0.36.1"
license: MIT
license_treatment: permissive
maintenance: active
---

# tiledb — Pythonic interface to the TileDB array storage manager

License: permissive · Maintenance: active · Downloads: 89.1K/mo

## What it is and what it does

TileDB-Py is a Python binding to the TileDB array storage engine, a columnar array database designed for efficient storage and retrieval of multi-dimensional data. It abstracts away the complexity of managing dense and sparse arrays, offering a NumPy-like interface for reading, writing, and querying array data. The package depends on numpy for numerical operations and packaging for version handling.

Typical workflows involve creating arrays with defined schemas, populating them with data via numpy arrays or Pandas DataFrames, and then querying slices or ranges efficiently. It targets scientific computing, geospatial analysis, and time-series workloads where traditional row-oriented databases are inefficient. The library is actively maintained, supports Python 3.9 through 3.13, and offers prebuilt wheels for major platforms.

Use it for:

- Store and query large geospatial raster datasets with efficient spatial indexing.
- Manage time-series data from sensors or financial markets with fast range queries.
- Build scientific computing pipelines that work with multi-dimensional arrays.
- Replace row-oriented storage when you need better compression and query performance.
- Integrate array data into machine learning workflows via Pandas DataFrames.

## Worth the install?

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

TileDB-Py provides a Python interface to the TileDB array storage engine, enabling you to read, write, and query multi-dimensional arrays with support for both dense and sparse data.

Yes, if you work with multi-dimensional array data and need efficient storage and querying. The active maintenance, permissive MIT license, and broad platform support make it a solid choice. Medium install friction is acceptable for the performance gains it offers. No known vulnerabilities as of the fact sheet date.

## Install

pip install tiledb
uv add tiledb
poetry add tiledb

## Installing tiledb

Before you install: Medium install friction: prebuilt wheels cover Python 3.10–3.13 on macOS (both ARM and x86), Linux (glibc 2.28+), and Windows, but the underlying TileDB C++ library must be present or bundled. Active maintenance with a recent release 170 days ago.

License in practice: MIT license is permissive; you may use, modify, and distribute tiledb freely in commercial and private projects with minimal restrictions.

Quickstart:

pip install tiledb
import tiledb
import numpy as np

# Create and write to an array
with tiledb.open('my_array.tdb', mode='w') as A:
    A[:] = np.arange(10)

Requires numpy and packaging at runtime; optional Pandas and PyArrow for dataframe functionality.

Verify before relying:

- Whether the prebuilt wheels include the full TileDB C++ runtime or require a separate system installation.
- Performance characteristics and scalability limits for typical array sizes and query patterns.
- Cloud storage backend support and any authentication requirements.

## Package facts

- License: MIT (permissive)
- Python support: supports_current
- Install friction: medium
- Maintenance: active
- Downloads: 89.1K/month (top 15,000 on PyPI)
- Known vulnerabilities: none known

## Tags

array storage engine python, multidimensional array database, tiledb python interface, sparse array storage, scientific array management, columnar array database, high-performance array queries, array-storage, scientific-computing, columnar-database

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