ray

Install

ray on PyPI

pip install ray

uv add ray

poetry add ray

Package facts

Evidence: ray-2.57.0-cp310-cp310-macosx_12_0_arm64.whl; ray-2.57.0-cp310-cp310-manylinux2014_aarch64.whl; ray-2.57.0-cp310-cp310-manylinux2014_x86_64.whl; ray-2.57.0-cp310-cp310-win_amd64.whl; ray-2.57.0-cp311-cp311-macosx_12_0_arm64.whl; ray-2.57.0-cp311-cp311-manylinux2014_aarch64.whl; ray-2.57.0-cp311-cp311-manylinux2014_x86_64.whl; ray-2.57.0-cp311-cp311-win_amd64.whl; ray-2.57.0-cp312-cp312-macosx_12_0_arm64.whl; ray-2.57.0-cp312-cp312-manylinux2014_aarch64.whl; ray-2.57.0-cp312-cp312-manylinux2014_x86_64.whl; ray-2.57.0-cp312-cp312-win_amd64.whl; ray-2.57.0-cp313-cp313-macosx_12_0_arm64.whl; ray-2.57.0-cp313-cp313-manylinux2014_aarch64.whl; ray-2.57.0-cp313-cp313-manylinux2014_x86_64.whl; ray-2.57.0-cp314-cp314-macosx_12_0_arm64.whl; ray-2.57.0-cp314-cp314-manylinux2014_aarch64.whl; ray-2.57.0-cp314-cp314-manylinux2014_x86_64.whl

Keywords: ray, distributed, parallel, machine-learning, hyperparameter-tuningreinforcement-learning, deep-learning, serving, python

About ray

from the package's own PyPI description — quoted content, verbatim

Read as markdown · JSON record · Source repository · Homepage

AI interpretation — verify before relying

AI-generated interpretation of the package facts above; every digit, version, license, or vulnerability id it cites is grounded in the facts already shown on this page

Ray is a distributed computing framework that scales Python applications from a single machine to clusters, providing abstractions for parallel tasks, stateful actors, and shared objects alongside ML libraries for data processing, training, hyperparameter tuning, and model serving.

Medium install friction due to 8 runtime dependencies (click, filelock, jsonschema, msgpack, packaging, protobuf, pyyaml, requests) and precompiled wheels for Python 3.10–3.14 across multiple platforms. Active maintenance with release 2.57.0 just 2 days old reduces adoption risk.

Apache 2.0 is permissive, allowing commercial use, modification, and distribution with minimal restrictions—suitable for most enterprise and open-source contexts.

pip install ray
import ray
ray.init()

@ray.remote
def hello():
    return "Hello from Ray"

result = ray.get(hello.remote())

Verdict: Ray is an actively maintained distributed computing framework with permissive Apache 2.0 licensing and no known vulnerabilities. Medium install friction is offset by broad platform support and established adoption. Suitable for scaling Python workloads from development to production, though cluster deployment adds operational complexity.

Needs verification

• Whether the 8 runtime dependencies introduce transitive security or stability concerns beyond OSV reports.
• Performance characteristics and resource overhead for small-scale or single-machine deployments.
• Compatibility guarantees across the supported Python 3.10–3.14 range in practice.
• Community adoption scale and production deployment prevalence.

Similar packages