Data

Tools

Indexes

Applications

Experiments

Summary

An open API service for producing an overview of a list of open source projects.

https://github.com/deepspeedai/DeepSpeed

billion-parameters compression data-parallelism deep-learning gpu inference machine-learning mixture-of-experts model-parallelism pipeline-parallelism pytorch trillion-parameters zero

Score: 34.01816764430561

Last synced: about 16 hours ago
JSON representation

Repository metadata:

DeepSpeed is a deep learning optimization library that makes distributed training and inference easy, efficient, and effective.

Owner metadata:

GitHub Events

Total
Last Year

Committers metadata

Last synced: 6 days ago

Total Commits: 3,308
Total Committers: 496
Avg Commits per committer: 6.669
Development Distribution Score (DDS): 0.866

Commits in past year: 425
Committers in past year: 111
Avg Commits per committer in past year: 3.829
Development Distribution Score (DDS) in past year: 0.755

Name Email Commits
Jeff Rasley j****y@m****m 444
Logan Adams 1****s 272
Olatunji Ruwase o****e@m****m 181
Masahiro Tanaka 8****a 178
Stas Bekman s****0 170
Michael Wyatt m****t@m****m 168
Shaden Smith S****h@m****m 96
Reza Yazdani 4****i 87
Ma, Guokai g****a@g****m 61
Ammar Ahmad Awan a****n@m****m 46
Conglong Li c****i@g****m 44
Nadav Elyahu 8****u 43
Lev Kurilenko 1****e 38
Joe Mayer 1****i 37
Olatunji Ruwase t****e@s****m 35
inkcherry m****u@i****m 33
Samyam Rajbhandari s****r@m****m 33
Yejing-Lai y****i@i****m 32
Cheng Li p****e@g****m 31
Connor Holmes c****s@m****m 31
Max Kovalenko m****o@h****i 27
Heyang Qin h****n@m****m 25
Ramya Ramineni 6****n 25
Molly Smith 1****h 24
digger yu d****u@o****m 23
Sam Ade Jacobs s****s@m****m 23
Quentin Anthony q****y@y****m 19
Liangliang Ma l****a@i****m 19
Raza Sikander 5****r 16
Guanhua Wang a****3@g****m 15
and 466 more...

Issue and Pull Request metadata

Last synced: 1 day ago

Total issues: 356
Total pull requests: 732
Average time to close issues: 2 months
Average time to close pull requests: 21 days
Total issue authors: 270
Total pull request authors: 170
Average comments per issue: 2.46
Average comments per pull request: 1.6
Merged pull request: 429
Bot issues: 18
Bot pull requests: 0

Past year issues: 104
Past year pull requests: 333
Past year average time to close issues: about 1 month
Past year average time to close pull requests: 7 days
Past year issue authors: 78
Past year pull request authors: 108
Past year average comments per issue: 1.92
Past year average comments per pull request: 1.95
Past year merged pull request: 186
Past year bot issues: 2
Past year bot pull requests: 0

More stats: https://issues.ecosyste.ms/repositories/lookup?url=https://github.com/deepspeedai/DeepSpeed

Top Issue Authors

  • github-actions[bot] (18)
  • hijkzzz (9)
  • janelu9 (7)
  • delock (7)
  • Flamefire (5)
  • fwerkor (4)
  • goransh-buh (4)
  • cyr0930 (4)
  • dujifeng (3)
  • tjruwase (3)
  • jamesbraza (3)
  • inkcherry (3)
  • alexk101 (3)
  • HollowMan6 (3)
  • LalchandPandia (3)

Top Pull Request Authors

  • loadams (95)
  • tohtana (71)
  • stas00 (45)
  • sfc-gh-truwase (30)
  • delock (23)
  • tjruwase (20)
  • nathon-lee (19)
  • inkcherry (18)
  • deepcharm (17)
  • hwchen2017 (15)
  • PKUWZP (15)
  • HollowMan6 (13)
  • vineethsaivs (13)
  • ebarkhordar (9)
  • jinyouzhi (9)

Top Issue Labels

  • bug (175)
  • training (127)
  • enhancement (54)
  • inference (25)
  • ci-failure (22)
  • deepspeed-chat (9)
  • compression (8)
  • install (3)
  • windows (3)
  • rocm (1)
  • compiler (1)
  • documentation (1)

Top Pull Request Labels

Package metadata

pypi.org: deepspeed

DeepSpeed library

  • Homepage: http://deepspeed.ai
  • Documentation: https://deepspeed.readthedocs.io
  • Licenses: Apache Software License 2.0
  • Latest release: 0.19.5 (published 15 days ago)
  • Last Synced: 2026-08-23T15:03:01.300Z (1 day ago)
  • Versions: 131
  • Dependent Packages: 109
  • Dependent Repositories: 3,724
  • Downloads: 1,040,214 Last month
  • Docker Downloads: 25,998,303
  • Rankings:
    • Stargazers count: 0.07%
    • Forks count: 0.1%
    • Dependent repos count: 0.165%
    • Dependent packages count: 0.186%
    • Average: 0.282%
    • Downloads: 0.529%
    • Docker downloads count: 0.64%
  • Maintainers (3)
  • Advisories:
proxy.golang.org: github.com/deepspeedai/DeepSpeed

proxy.golang.org: github.com/deepspeedai/deepspeed

anaconda.org: deepspeed

DeepSpeed is a deep learning optimization library that makes distributed training and inference easy, efficient, and effective. It enables world-class training speed, memory efficiency, scalability, and accessibility for large model training and inference.

  • Homepage: https://www.deepspeed.ai
  • Licenses: Apache-2.0
  • Latest release: 0.18.9 (published 4 months ago)
  • Last Synced: 2026-05-06T14:51:24.588Z (4 months ago)
  • Versions: 1
  • Dependent Packages: 0
  • Dependent Repositories: 0
  • Downloads: 15 Total
  • Rankings:
    • Dependent packages count: 35.874%
    • Dependent repos count: 39.116%
    • Average: 47.648%
    • Downloads: 67.954%

Dependencies

.github/workflows/amd-mi200.yml actions
  • ./.github/workflows/setup-venv * composite
  • JasonEtco/create-an-issue v2 composite
  • actions/checkout v3 composite
.github/workflows/formatting.yml actions
  • actions/checkout v3 composite
.github/workflows/nv-accelerate-v100.yml actions
  • ./.github/workflows/setup-venv * composite
  • actions/checkout v3 composite
.github/workflows/nv-ds-chat.yml actions
  • ./.github/workflows/setup-venv * composite
  • JasonEtco/create-an-issue v2 composite
  • actions/checkout v3 composite
.github/workflows/nv-inference.yml actions
  • ./.github/workflows/setup-venv * composite
  • actions/checkout v3 composite
.github/workflows/nv-lightning-v100.yml actions
  • ./.github/workflows/setup-venv * composite
  • actions/checkout v3 composite
.github/workflows/nv-mii.yml actions
  • ./.github/workflows/setup-venv * composite
  • actions/checkout v3 composite
.github/workflows/nv-nightly.yml actions
  • ./.github/workflows/setup-venv * composite
  • JasonEtco/create-an-issue v2 composite
  • actions/checkout v3 composite
.github/workflows/nv-pre-compile-ops.yml actions
  • actions/checkout v3 composite
.github/workflows/nv-torch-latest-v100.yml actions
  • ./.github/workflows/setup-venv * composite
  • actions/checkout v3 composite
.github/workflows/nv-torch-nightly-v100.yml actions
  • ./.github/workflows/setup-venv * composite
  • JasonEtco/create-an-issue v2 composite
  • actions/checkout v3 composite
.github/workflows/nv-transformers-v100.yml actions
  • ./.github/workflows/setup-venv * composite
  • actions/checkout v3 composite
.github/workflows/python.yml actions
  • actions/checkout v3 composite
.github/workflows/setup-venv/action.yml actions
requirements/requirements-triton.txt pypi
  • triton ==2.1.0
.github/workflows/modal-torch-latest.yml actions
  • actions/checkout v7 composite
  • actions/download-artifact v4 composite
  • actions/setup-python v7 composite
  • actions/upload-artifact v4 composite
requirements/requirements-1bit-mpi.txt pypi
  • mpi4py *
requirements/requirements-dev.txt pypi
  • accelerate * development
  • clang-format ==18.1.3 development
  • comet_ml >=3.41.0 development
  • deepspeed-kernels * development
  • docutils <0.18 development
  • future * development
  • importlib-metadata >=4 development
  • mup * development
  • pre-commit >=3.2.0 development
  • pytest >=7.2.0,<8.4.0 development
  • pytest-forked * development
  • pytest-randomly * development
  • pytest-xdist * development
  • qtorch ==0.3.0 development
  • recommonmark * development
  • sphinx * development
  • sphinx-rtd-theme * development
  • tensorboard * development
  • torchvision * development
  • transformers >=4.51.3 development
  • wandb * development
requirements/requirements-sparse_attn.txt pypi
  • triton ==1.0.0
requirements/requirements-autotuning-ml.txt pypi
  • hjson *
  • tabulate *
  • xgboost *
environment.yml conda
  • ca-certificates
  • certifi
  • compilers
  • cuda
  • gcc 11.4
  • libaio
  • ninja
  • openssl
  • py-cpuinfo
  • pydantic
  • python 3.10
  • pytorch
  • pytorch-cuda 11.8
  • sysroot_linux-64 2.17
  • torchaudio
  • torchvision
.github/workflows/aws-torch-latest-full.yml actions
  • actions/checkout v4 composite
.github/workflows/aws-accelerate.yml actions
  • actions/checkout v7 composite
  • actions/checkout v4 composite
  • dorny/paths-filter v4 composite
.github/workflows/dco.yml actions
.github/workflows/nv-sd.yml actions
  • JasonEtco/create-an-issue v2 composite
  • actions/checkout v4 composite
.github/workflows/no-torch.yml actions
  • ./.github/workflows/setup-venv * composite
  • JasonEtco/create-an-issue v2 composite
  • actions/checkout v7 composite
.github/workflows/cpu-torch-latest.yml actions
  • ./.github/workflows/setup-venv * composite
  • actions/checkout v7 composite
  • dorny/paths-filter v4 composite
.github/workflows/nv-a6000.yml actions
  • actions/checkout v4 composite
requirements/requirements-sparse_pruning.txt pypi
  • neural-compressor ==2.1.0
.github/workflows/xpu-max1100.yml actions
  • actions/checkout v4 composite
requirements/requirements-readthedocs.txt pypi
  • autodoc_pydantic >=2.0.0
  • docutils <0.18
  • hjson *
  • packaging *
  • psutil *
  • py-cpuinfo *
  • pydantic >=2.0.0
  • recommonmark *
  • sphinx_rtd_theme *
  • torch >=2.0.0
  • tqdm *
requirements/requirements-sd.txt pypi
  • diffusers >=0.25.0
  • triton >=2.1.0
docker/rocm/Dockerfile docker
  • rocm/pytorch latest build
requirements/requirements-autotuning.txt pypi
  • tabulate *
.github/workflows/nv-flash-attn.yml actions
  • JasonEtco/create-an-issue v2 composite
  • actions/checkout v4 composite
docker/Dockerfile docker
  • nvidia/cuda 12.2.2-devel-ubuntu20.04 build
requirements/requirements-cpu.txt pypi
requirements/requirements-inf.txt pypi
  • google *
  • lm-eval ==0.3.0
  • protobuf *
  • qtorch *
  • safetensors *
  • sentencepiece *
  • transformers >=4.32.1
.github/workflows/aws-torch-latest.yml actions
  • actions/checkout v7 composite
  • actions/checkout v4 composite
  • dorny/paths-filter v4 composite
.github/workflows/modal-accelerate.yml actions
  • actions/checkout v7 composite
  • actions/setup-python v7 composite
  • dorny/paths-filter v4 composite
.github/workflows/xpu-compile.yml actions
  • actions/checkout v4 composite
.github/workflows/hpu-gaudi2-nightly.yml actions
  • actions/checkout v4 composite
.github/workflows/hpu-gaudi2.yml actions
  • actions/checkout v4 composite
.github/workflows/release.yml actions
  • ./.github/workflows/setup-venv * composite
  • actions/checkout v7 composite
  • peter-evans/create-pull-request v8 composite
  • pypa/gh-action-pypi-publish release/v1 composite
requirements/requirements-deepcompile.txt pypi
  • scipy *
requirements/requirements.txt pypi
  • einops *
  • hjson *
  • msgpack *
  • ninja *
  • numpy *
  • packaging >=20.0
  • psutil *
  • py-cpuinfo *
  • pydantic >=2.0.0
  • torch >=2.0.0
  • tqdm *
requirements/requirements-torchembed.txt pypi
  • torchembed >=0.3.1
setup.py pypi
docs/Gemfile rubygems
  • github-pages >= 0
  • jekyll-feed >= 0
  • jekyll-include-cache >= 0
  • jekyll-paginate >= 0
  • jekyll-remote-theme >= 0
  • minimal-mistakes-jekyll >= 0
  • tzinfo ~> 1.2
  • tzinfo-data >= 0
  • wdm ~> 0.1.1
  • webrick ~> 1.8