clearml/clearml

What is it?

Key features

• 01Automatic logging of code state, environment, hyperparameters and console output for each experiment
• 02Dataset versioning and management on S3, GCS, Azure and NAS storage
• 03Remote job execution through agents on cloud, Kubernetes or bare-metal machines
• 04Pipeline and task orchestration with scheduling and queues
• 05Model serving with GPU support and built-in monitoring

Use cases

• Tracking and comparing deep learning experiments across a team
• Running training jobs on a shared GPU cluster or cloud instances
• Versioning large datasets stored in object storage
• Building automated training pipelines from existing experiments

Quick start

1. 1

   Sign up for the free ClearML Hosted Service or set up your own server

2. 2

   Install the clearml Python package

   pip install clearml
3. 3

   Connect the SDK to the server by creating credentials and following the prompts

   clearml-init
4. 4

   Add two lines to your code to start automatic logging

Things to know

• !Full functionality relies on a ClearML server, either the hosted service or one you deploy yourself.
• !Self-hosting the server requires extra setup effort.
• !The public demo server is no longer used by default and experiments sent to it are public.

Topics

FAQ

What is clearml?

Open-source MLOps/LLMOps suite covering experiment tracking, dataset versioning, pipelines, orchestration, scheduling and model serving. ClearML is a suite for managing machine learning and deep learning workflows. Its experiment manager captures a run's source control state, package environment, hyperparameters, console output, scalars and artifacts after adding two lines of code to a Python script. Data management lets teams version datasets on object storage such as S3, GCS, Azure or NAS.

Is clearml free to use? What license does it have?

clearml is open source under the Apache-2.0 license. Check the license terms before using it in commercial products.

Is clearml still maintained?

Actively maintained. The last commit to clearml was on Nov 3, 2026.

What can I use clearml for?

Tracking and comparing deep learning experiments across a team Running training jobs on a shared GPU cluster or cloud instances Versioning large datasets stored in object storage Building automated training pipelines from existing experiments

How do I install clearml?

Sign up for the free ClearML Hosted Service or set up your own server Install the clearml Python package: pip install clearml Connect the SDK to the server by creating credentials and following the prompts: clearml-init Add two lines to your code to start automatic logging

Where can I get clearml?

The source code and installation instructions are on GitHub: https://github.com/clearml/clearml