#99th of 22 in ML frameworks and libraries
ray
Distributed Python runtime with libraries for data, training, tuning, RL and serving
Documentation ↗
(opens in a new tab)Website ↗
(opens in a new tab)Repository on GitHub ↗
(opens in a new tab)
- Stars
- 44k
- License
- Apache-2.0
- Last commit
- Oct 2026
- Last release
- Oct 2026
- Language
- Python
41,497 on 25 Mar 202644,031 on 14 Oct 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
Ray is a distributed runtime for scaling Python code from a laptop to a cluster, built around tasks, actors and shared objects. On top of the core it ships libraries for datasets (Data), distributed training (Train), hyperparameter tuning (Tune), reinforcement learning (RLlib) and model serving (Serve). It includes a dashboard and a distributed debugger, and runs on local machines, clusters, cloud providers and Kubernetes.
Who it is for: Python and ML engineers scaling workloads across machines or clusters
Strengths
- Same Python code scales from one machine to a cluster without extra infrastructure
- One runtime covers data processing, training, tuning, RL and serving
- Runs on bare clusters, cloud providers and Kubernetes
- Ships a dashboard and a distributed debugger for cluster apps
Weaknesses
- Python-centric; README describes scaling Python applications only
- Broad scope with many libraries means a large surface to learn
- README gives no hardware requirements or cluster sizing guidance
- Anyscale managed offering is promoted; OSS versus hosted feature split is unclear
What it needs
- GPU optional
- Docker
Also in ML frameworks and libraries
See all 22| Rank | Project | Score |
|---|---|---|
| 1 | transformersModel-definition library for text, vision, audio and multimodal models | 94 out of 100 |
| 2 | pytorchPython tensor library with GPU support and tape-based autograd for deep learning | 93 out of 100 |
| 3 | ultralyticsPython library and CLI for training and running YOLO vision models | 88 out of 100 |
| #4 | supervisionPython toolkit for loading, annotating and counting computer vision detections | 84 out of 100 |
| #5 | pytorch-image-modelsPyTorch library of image models, pretrained weights and reference training scripts | 82 out of 100 |