#6666th of 126 in ML frameworks and libraries
clearml
Experiment tracking, pipelines and dataset versioning for ML jobs
- Stars
- 6.9k
- License
- Apache-2.0
- Last commit
- Oct 2026
- Last release
- Aug 2026
- Language
- Python
6,568 on 4 Apr 20266,927 on 31 Oct 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
ClearML is a Python SDK plus server for tracking ML experiments, running remote jobs through agents, and versioning datasets on object storage. Adding Task.init() to a script logs git state, packages, hyperparameters, console output, resource use and framework metrics. Pipelines, hyperparameter optimization, model serving and an orchestration dashboard are separate modules.
Who it is for: ML teams tracking experiments, datasets and remote training jobs
Strengths
- Two-line Task.init() captures git state, packages, parameters, console output and GPU/CPU metrics
- Auto-logs PyTorch, TensorFlow, Keras, XGBoost, LightGBM, scikit-learn, Matplotlib and TensorBoard
- Dataset versioning on S3, GS, Azure or NAS, with a clearml-data CLI
- Agents run jobs on Kubernetes, cloud or bare metal; pipelines and hyperparameter optimization included
Weaknesses
- Full use needs a ClearML server, hosted or self-deployed
- Serving, agent, session and fractional GPU modules live in separate repositories
- Centered on classic ML/DL training; LLM features are only briefly described
- README gives no RAM, port or server dependency details
What it needs
- GPU optional
- Needs ClearML Server (hosted or self-hosted), clearml-agent, S3/GS/Azure/NAS (optional storage)
- Models: PyTorch, TensorFlow, Keras, FastAI, XGBoost
Quick answers
Is clearml open source?
Yes. Its code is on GitHub under the Apache-2.0 license.
Does clearml need a GPU?
No. It runs on a CPU and can use a GPU when you have one.
Also in ML frameworks and libraries
See all 126| Rank | Project | Score |
|---|---|---|
| 1 | pytorchPython tensor library with GPU support and tape-based autograd for deep learning | 95 out of 100 |
| 2 | transformersPython library defining open models for text, vision, audio and multimodal tasks | 94 out of 100 |
| 3 | ultralyticsPython library and CLI for training and running YOLO vision models | 93 out of 100 |
| #4 | pytorch-image-modelsPyTorch library of image models, pretrained weights and reference training scripts | 93 out of 100 |
| #5 | supervisionPython toolkit for loading, annotating and counting computer vision detections | 91 out of 100 |