Polars Nanook

The compute platform for all your workloads

A high-performance platform at any scale, built for open data formats and ready for your agents.

Trusted by data and AI teams around the world

Benefits

Four reasons organizations build with Nanook

Write once, run anywhere

The same Polars code you write locally runs in production, exactly as it is. The scale changes. Your code doesn't.

Scale without complexity

Grow from a single machine to a full cluster exactly when you need to, without taking on a distributed-systems project of your own.

Optimize and profile your queries

Nanook comes with an advanced query profiler and built-in MCP server, allowing you to reason about your query performance and keep optimizing.

Data stays in your environment

Your data stays exactly where it is, inside your own infrastructure. All we ever see is the metadata that powers your dashboard.

Infrastructure

Runs in your infrastructure, next to your data

Deploy on AWS, Kubernetes, Ray or bare metal. With your data and compute in the same place, queries are fast and nothing leaves your network.

  • AzureComing soon
  • GCPComing soon
Row of infrastructure tiles, Azure, Ray, Kubernetes, AWS and Google Cloud, glowing over a dark aurora
On-premise

Deploy entirely in your own infrastructure

The same scalable performance as Nanook in the cloud, with zero external dependencies. Your data stays yours, always.

Configuration panel with cluster size, CPU and memory options
Infrastructure

Integrated with your infrastructure

Polars fits into the stack you already use, from on-premises hardware to your existing cloud, data platforms, and tools.

A sealed server rack glowing on its own, apart from the racks around it
Environment

Air-gapped ready

Nanook is built to run in fully isolated environments, sealed off from the outside entirely.

Shield marked SOC II
Compliance

Meets your compliance standards

Polars is SOC 2 audited, and your data never leaves your environment, so your security and compliance teams stay in control.

Case studies

Case studies

A cornerstone for advanced analytics at scale

“Polars has become a cornerstone for advanced analytics, delivering exceptional efficiency on even the most complex transformations.”

Derek CaramellaVP, Enterprise Marketing Data Analytics @ Citizens
Citizens wordmark over a blue and teal aurora
Features

Built for data teams,
from development to production

Query visibility

Real-time observability into every query and every cluster, so you can always see exactly what's running and why.

The Polars bear tile surrounded by Codex, Copilot, Claude, ChatGPT, Cursor and OpenCode tiles

Built for agents

Your agents connect over MCP to find any query, see how it ran stage by stage, and suggest how to make it faster.

Extensible

Add your own custom expressions, just as fast and easy to use as the ones we ship.

Integrations

Fits the infrastructure you already run

Run small jobs on a single node and large ones across a cluster, from one platform that connects to your storage, infrastructure and observability tools.

Storage
Formats
Catalogs
Deploy
Observe
Accelerator
Request a demo

Demo Nanook

See Polars at scale, and talk through deployment (managed, on-premises, or air-gapped), pricing, and support. A specialist will get back to you.

  • Scale without complexity

    Grow from a single machine to a full cluster.

  • Data stays in your environment

    Your data stays exactly where it is, inside your own infrastructure.

All fields and the privacy box are required.

Sending your request.

Something went wrong on our side. Try again in a moment.

Request received

A member of the Polars team will be in touch soon.

Learn

Learn resources

FAQs

Frequently asked questions

Polars Nanook is free to start, with 2,000 vCPU-hours a month on Developer. Beyond that, you pay as you go per vCPU-hour, and organizations with large workloads can reach out for a volume discount on Team or custom pricing on Enterprise.

Learn more