---
id: dask-expr
version: "2.0.0"
license: BSD
license_treatment: permissive
maintenance: dormant
---

# dask-expr — High Level Expressions for Dask

License: permissive · Maintenance: dormant · Downloads: 11.0M/mo

## What it is and what it does

Dask Expressions is a rewrite of Dask DataFrame that adds query optimization by representing user operations as an expression tree before execution. Instead of executing operations immediately, the library builds a tree structure encoding the computation, optimizes it (e.g., fusing operations, reordering steps), and then executes the optimized plan. It is designed as a drop-in replacement and has been the default backend for dask.DataFrame since version 2024.3.0.

The package depends only on dask and installs with low friction. It covers nearly the full Dask DataFrame API, with the exception of named GroupBy Aggregations. It supports modern Python versions (3.10 through 3.13) and is intended for developers and researchers working with distributed data processing at scale.

Use it for:

- Optimize large distributed dataframe queries by leveraging automatic expression tree optimization before execution.
- Migrate from standard Dask DataFrame to a more efficient backend without changing existing code.
- Process time-series and grouped data with better performance through query planning.
- Build data pipelines where query optimization can reduce computation time and memory usage.
- Work with pandas-like APIs on distributed datasets while benefiting from automatic optimization.

## Worth the install?

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

Dask Expressions provides query optimization for Dask DataFrames by encoding operations as an expression tree that is optimized before execution, replacing the earlier Dask DataFrame implementation.

Yes. Dask Expressions is the default backend for dask.DataFrame as of version 2024.3.0, making it the standard choice for new Dask DataFrame work. It offers query optimization with low install friction and permissive licensing. Maintenance is dormant but the package is stable and widely used. Install it if you are using Dask DataFrames; it is the recommended path forward.

## Install

pip install dask-expr
uv add dask-expr
poetry add dask-expr

## Installing dask-expr

Before you install: Low install friction with a single wheel distribution. Maintenance is dormant (570 days since last release), though the repository remains active with a recent commit on 2025-01-21. This is the default backend for dask.DataFrame since version 2024.3.0, so stability is established.

License in practice: BSD license is permissive, allowing commercial use, modification, and distribution with minimal restrictions—suitable for most projects.

Quickstart:

pip install dask-expr

import dask_expr as dx

df = dx.datasets.timeseries()
result = df.groupby("name").x.mean().compute()

Requires Python 3.10 or later.

Verify before relying:

- Whether named GroupBy Aggregations (noted as missing) are critical for your workflow.
- Performance gains over standard Dask DataFrame for your specific query patterns.

## Package facts

- License: BSD (permissive)
- Python support: supports_current
- Install friction: low
- Maintenance: dormant
- Downloads: 11.0M/month (top 5,000 on PyPI)
- Known vulnerabilities: none known

## Tags

dask dataframe query optimization, distributed dataframe with optimization, dask expression trees, optimized pandas-like operations, dask dataframe backend, lazy dataframe evaluation, distributed data processing, query-optimization, distributed-computing, dataframe

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