项目文件夹

T

Marwan Sarieddine

[Train] Retry py-spy with sudo in the NCCL hang detector's stack dump (#67794)

880a21ecfd6小时前提交

[release] Remove the ray-ml release test image and the gpu byod type (4/N) (#67518 )

19小时前

[Core/Auth] Add token auth rules to security.md and Bugbot (#67726 )

昨天

[Core/Auth] Add token auth rules to security.md and Bugbot (#67726 )

昨天

[Chore] Disable gemini code review bot for draft PR (#64862 )

4个月前

[ci] Request ray-ci review on all Dependabot PRs (#67306 )

2周前

[doc][tune] Seed the Ray Tune Vale term library (warn-first, not yet enforced) (#67017 )

2周前

[Core/Auth] Make token auth the default for all clusters (#67523 )

前天

[release] Remove the ray-ml release test image and the gpu byod type (4/N) (#67518 )

19小时前

[Core/Auth] Make token auth the default for all clusters (#67523 )

前天

[Core/Auth][Spark] Create the auth token in Ray on Spark tests and document token setup for Ray on Spark (#67796 )

13小时前

[Core/Auth] Make token auth the default for all clusters (#67523 )

前天

[Core/Auth] Make token auth the default for all clusters (#67523 )

前天

[Train] Retry py-spy with sudo in the NCCL hang detector's stack dump (#67794 )

6小时前

[release] Move mid_epoch_resume_benchmark to the torch-cu128 BYOD image (#67805 )

16小时前

[core] Restrict an existing default auth token file and ~/.ray to their owner on load (#67776 )

14小时前

[core][build] Drop -Wl,-pie from vendored RocksDB WITH_TSAN link flags (#64917 )

4个月前

[build] State --enable_workspace explicitly in .bazelrc (#66305 )

2周前

[Build] Upgrade Bazel from 6.5.0 to 7.5.0 (#61601 )

8个月前

[Lint] One parameter/argument per line for C++ code (#22725 )

4年前

[core] Lint cpp files in core_worker (#48789 )

2年前

Fix editorconfig option name (#50993 )

去年

Move `/rllib/` into `/python/ray/` (#66096 )

上个月

.git-blame-ignore-revs

[doc] Ignore the RLlib soft-wrap reflow commit in git blame (#66285 )

2个月前

[Core] Auto generate protobuf and grpc code (#40663 )

3年前

[doc] add agent context for Ray documentation (#63227 )

6个月前

.pre-commit-config.yaml

[Data][CI] Run the Ray Data pyrefly check in pre-commit (#67652 )

昨天

.prettierrc.toml

[Doc] Fix broken splash.html (#42029 )

3年前

[ci] Gate the docstring-only and docs-example-test labels with rayci label rules (#67341 )

上周

.readthedocs.yaml

[ci][doc] Let a PR opt into the Read the Docs preview with an @rtd-build commit line (#66805 )

3周前

[doc][tune] Seed the Ray Tune Vale term library (warn-first, not yet enforced) (#67017 )

2周前

[ci] add docker/base-deps, base-extra, ray-llm to .wandaspecs (#61450 )

9个月前

[docs] AGENTS.md: add redirects for API renames that move docs URLs (#67240 )

2周前

build-docker.sh

[ci] removing requirement file and constraint file build args (#60151 )

10个月前

[build] feat: add top-level build-image.sh (#61042 )

9个月前

[ci] refactor: extract shared build utilities into build_common.py (#61139 )

9个月前

[core] Remove gmock dependency (#66538 )

上个月

CONTRIBUTING.md

Improve `contributors.rst` (#67116 )

2周前

gen_py_proto.py

[wheel] add `gen_py_proto` script (#55957 )

去年

[wheel] hermetic wheel building (#55091 )

去年

gen_redis_pkg.py

add gen_redis_pkg bazel target (#56527 )

去年

[Core] Vendor setproctitle (#53471 )

去年

RLLIB and pylintrc (#8995 )

6年前

[ci] Bump ruff pre-commit hook from v0.8.4 to v0.17.3 (#67609 )

前天

[Data] Add HiveServer2-backed read_hive datasource (#66573 )

3周前

[train] Add missing `@pytest.mark.asyncio` to async def tests (#61849 )

8个月前

ray-images.json

[ci] Build and publish cu133 (CUDA 13.3.1) images + hello_world release test (#66547 )

上周

[ci] port Buildkite Ray Images to Wanda (#60455 )

9个月前

[ci][doc] Convert the root README and other READMEs from RST to Markdown (#66811 )

前天

[Doc] Fix security doc link (#46352 )

2年前

Use git's rev-parse to get absolute paths. (#64513 )

4个月前

[build] Move rules_python off 0.9.0 (#66207 )

2个月前

README.md

Ray is a unified framework for scaling AI and Python applications. Ray consists of a core distributed runtime and a set of AI libraries for simplifying ML compute:

Learn more about Ray AI Libraries:

Data: Scalable Datasets for ML
Train: Distributed Training
Tune: Scalable Hyperparameter Tuning
RLlib: Scalable Reinforcement Learning
Serve: Scalable and Programmable Serving

Or more about Ray Core and its key abstractions:

Tasks: Stateless functions executed in the cluster.
Actors: Stateful worker processes created in the cluster.
Objects: Immutable values accessible across the cluster.

Learn more about Monitoring and Debugging:

Monitor Ray apps and clusters with the Ray Dashboard.
Debug Ray apps with the Ray Distributed Debugger.

Ray runs on any machine, cluster, cloud provider, and Kubernetes, and features a growing ecosystem of community integrations.

Install Ray with: pip install ray. For nightly wheels, see the Installation page.

Why Ray?

Today's ML workloads are increasingly compute-intensive. As convenient as they are, single-node development environments such as your laptop cannot scale to meet these demands.

Ray is a unified way to scale Python and AI applications from a laptop to a cluster.

With Ray, you can seamlessly scale the same code from a laptop to a cluster. Ray is designed to be general-purpose, meaning that it can performantly run any kind of workload. If your application is written in Python, you can scale it with Ray, no other infrastructure required.

More Information

Older documents:

Getting Involved

Platform	Purpose	Estimated Response Time	Support Level
Discourse Forum	For discussions about development and questions about usage.	< 1 day	Community
GitHub Issues	For reporting bugs and filing feature requests.	< 2 days	Ray OSS Team
Slack	For collaborating with other Ray users.	< 2 days	Community
StackOverflow	For asking questions about how to use Ray.	3-5 days	Community
Meetup Group	For learning about Ray projects and best practices.	Monthly	Ray DevRel
Twitter	For staying up-to-date on new features.	Daily	Ray DevRel