---
id: coiled
version: "1.135.3"
license: Elastic-2.0
license_treatment: unclear
maintenance: active
---

# coiled — Python client for coiled.io dask clusters

License: unclear · Maintenance: active · Downloads: 1.4M/mo

## What it is and what it does

Coiled is a client library that bridges your local Python environment to the coiled.io deployment service, which provisions and manages Dask clusters in the cloud. It abstracts away the complexity of setting up cloud infrastructure, configuring networking, and managing software environments—you call Coiled's API to create a cluster, and the service handles resource allocation and orchestration on your behalf.

The package depends on dask and distributed for the core parallel computing engine, boto3 for AWS integration, fabric and paramiko for remote execution, and several utility libraries for configuration and monitoring. It's designed for teams running large-scale analytics or machine learning workloads that need to scale beyond a single machine without manually provisioning and configuring cloud infrastructure.

Use it for:

- Scale a Dask DataFrame operation across multiple cloud instances without manually provisioning VMs or configuring networking.
- Run distributed machine learning training jobs that require more compute than your local machine provides.
- Provision temporary compute clusters for batch analytics jobs, then tear them down to control costs.
- Integrate cloud-based Dask clusters into a CI/CD pipeline for large-scale data processing.
- Manage multiple Dask clusters across different cloud regions or providers from a single Python client.

## Worth the install?

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

Coiled is a Python client library that provisions and manages Dask clusters on cloud infrastructure, handling resource allocation, networking, and environment setup to scale parallel workloads.

Yes, if you use Dask and need to scale to cloud infrastructure without managing infrastructure yourself. The low install friction and active maintenance are positive signals. However, verify the Elastic-2.0 license terms for your use case, and confirm whether coiled.io's service requires a paid account—the client library alone is not sufficient without access to the backend service.

## Install

pip install coiled
uv add coiled
poetry add coiled

## Installing coiled

Before you install: Low install friction with a pure-Python wheel. Active maintenance as of the release date. The 24 runtime dependencies include substantial libraries like dask, distributed, boto3, and fabric, which may add complexity to your environment but are standard for cloud-based distributed computing.

License in practice: Licensed under Elastic-2.0, but license treatment is marked unclear in the metadata. You should verify the exact terms before using in proprietary or commercial contexts, as Elastic licenses can carry specific restrictions.

Quickstart:

pip install coiled
import coiled
cluster = coiled.Cluster()
cluster.close()

Requires Python 3.9 or later. Coiled.io account and cloud credentials (AWS or other supported provider) needed to actually provision clusters.

Verify before relying:

- Whether Elastic-2.0 license permits commercial use without restrictions or additional agreements.
- Whether the coiled.io service itself requires a paid subscription beyond the open-source client library.

## Package facts

- License: Elastic-2.0 (unclear)
- Python support: supports_current
- Install friction: low
- Maintenance: active
- Downloads: 1.4M/month (top 5,000 on PyPI)
- Known vulnerabilities: none known

## Tags

dask cluster deployment, python distributed computing cloud, scale dask workloads, managed dask clusters, cloud parallel analytics, dask cluster provisioning, serverless dask scaling, dask-integration, cloud-provisioning, distributed-computing

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