Data

Tools

Indexes

Applications

Experiments

Summary

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

https://github.com/huggingface/accelerate

Score: 32.91693644020296

Last synced: about 21 hours ago
JSON representation

Repository metadata:

🚀 A simple way to launch, train, and use PyTorch models on almost any device and distributed configuration, automatic mixed precision (including fp8), and easy-to-configure FSDP and DeepSpeed support

Owner metadata:

GitHub Events

Total
Last Year

Committers metadata

Last synced: 2 days ago

Total Commits: 2,078
Total Committers: 455
Avg Commits per committer: 4.567
Development Distribution Score (DDS): 0.732

Commits in past year: 219
Committers in past year: 103
Avg Commits per committer in past year: 2.126
Development Distribution Score (DDS) in past year: 0.831

Name Email Commits
Zachary Mueller m****r@g****m 557
Sylvain Gugger s****r@g****m 285
Marc Sun 5****c 145
Sourab Mangrulkar 1****0 116
Fanli Lin f****n@i****m 51
Yao Matrix m****o@i****m 27
Younes Belkada 4****a 22
Huazhong Ji h****0@g****m 21
Matej Sirovatka 5****1 19
Benjamin Bossan B****n 17
Quentin Gallouédec 4****c 17
Wang, Yi y****g@i****m 17
Abhilash Majumder 3****0 14
Stas Bekman s****0 14
Kashif Rasul k****l@g****m 11
Steven Liu 5****u 11
Ilyas Moutawwakil 5****l 10
fxmarty 9****y 10
Thomas Wolf t****f 10
LiamSwayne 1****e 10
jiqing-feng j****g@i****m 10
YH y****a@l****o 9
Michael Benayoun m****l@h****o 9
dependabot[bot] 4****] 8
Aarni Koskela a****x@i****i 7
Patrick von Platen p****n@g****m 7
Mehant Kammakomati m****2@i****m 7
YU XINYUAN x****j@c****k 7
Yuanyuan Chen c****r@o****m 6
Yu Chin Fabian Lim f****m 6
and 425 more...

Issue and Pull Request metadata

Last synced: 2 days ago

Total issues: 1,308
Total pull requests: 1,850
Average time to close issues: about 1 month
Average time to close pull requests: 11 days
Total issue authors: 1,035
Total pull request authors: 481
Average comments per issue: 3.83
Average comments per pull request: 2.2
Merged pull request: 1,221
Bot issues: 0
Bot pull requests: 23

Past year issues: 70
Past year pull requests: 336
Past year average time to close issues: about 1 month
Past year average time to close pull requests: 14 days
Past year issue authors: 59
Past year pull request authors: 177
Past year average comments per issue: 1.17
Past year average comments per pull request: 1.03
Past year merged pull request: 93
Past year bot issues: 0
Past year bot pull requests: 19

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

Top Issue Authors

  • muellerzr (13)
  • sayakpaul (9)
  • bghira (9)
  • stas00 (7)
  • pratogab (6)
  • fancyerii (6)
  • npuichigo (6)
  • Hannibal046 (5)
  • dineshkh (5)
  • jubueche (5)
  • ojh31 (4)
  • efsotr (4)
  • winglian (4)
  • GabPrato (4)
  • nebrelbug (4)

Top Pull Request Authors

  • muellerzr (341)
  • SunMarc (136)
  • faaany (94)
  • pacman100 (35)
  • S1ro1 (30)
  • yao-matrix (29)
  • statelesshz (26)
  • BenjaminBossan (23)
  • qgallouedec (22)
  • fxmarty (20)
  • dependabot[bot] (20)
  • yhna940 (20)
  • stevhliu (19)
  • vineethsaivs (18)
  • akx (18)

Top Issue Labels

  • solved (43)
  • feature request (38)
  • enhancement (30)
  • bug (19)
  • wip (14)
  • documentation (8)
  • contributions-welcome (3)
  • good first issue (3)
  • wontfix (3)
  • big model inference (2)
  • terminal (1)
  • info-needed (1)
  • pippy (1)
  • NPU (1)
  • GPU (1)
  • deepspeed (1)
  • TPU (1)

Top Pull Request Labels

  • dependencies (20)
  • github_actions (16)
  • enhancement (14)
  • wip (13)
  • code agent slop (10)
  • bug (5)
  • feature request (2)
  • documentation (2)
  • for patch (2)
  • TPU (2)
  • deepspeed (2)
  • GPU (1)
  • area/security (1)
  • per-call-memory/neutral (1)
  • bloat/none (1)
  • lang/c++ (1)
  • CLA Signed (1)
  • Stale (1)
  • React Core Team (1)
  • go (1)
  • lang/core (1)
  • per-channel-memory/neutral (1)
  • imported (1)
  • release notes: no (1)

Package metadata

pypi.org: accelerate

Accelerate

proxy.golang.org: github.com/huggingface/accelerate

  • Homepage:
  • Documentation: https://pkg.go.dev/github.com/huggingface/accelerate#section-documentation
  • Licenses: apache-2.0
  • Latest release: v1.15.0 (published about 2 months ago)
  • Last Synced: 2026-10-31T15:02:20.724Z (2 days ago)
  • Versions: 79
  • Dependent Packages: 0
  • Dependent Repositories: 0
  • Rankings:
    • Stargazers count: 0.871%
    • Forks count: 1.143%
    • Average: 5.261%
    • Dependent packages count: 8.384%
    • Dependent repos count: 10.647%
conda-forge.org: accelerate

  • Homepage: https://github.com/huggingface/accelerate
  • Licenses: Apache-2.0
  • Latest release: 0.14.0 (published almost 4 years ago)
  • Last Synced: 2026-10-24T10:29:36.125Z (9 days ago)
  • Versions: 13
  • Dependent Packages: 1
  • Dependent Repositories: 11
  • Rankings:
    • Stargazers count: 5.739%
    • Forks count: 8.703%
    • Dependent repos count: 10.669%
    • Average: 13.521%
    • Dependent packages count: 28.974%
spack.io: py-accelerate

A simple way to train and use PyTorch models with multi-GPU, TPU, mixed- precision.

  • Homepage: https://github.com/huggingface/accelerate
  • Licenses: []
  • Latest release: 1.12.0 (published 3 months ago)
  • Last Synced: 2026-10-31T15:02:28.677Z (2 days ago)
  • Versions: 4
  • Dependent Packages: 1
  • Dependent Repositories: 0
  • Rankings:
    • Dependent repos count: 0.0%
    • Stargazers count: 2.696%
    • Forks count: 5.252%
    • Average: 16.305%
    • Dependent packages count: 57.274%
  • Maintainers (1)
anaconda.org: huggingface_accelerate

Accelerate was created for PyTorch users who like to write the training loop of PyTorch models but are reluctant to write and maintain the boilerplate code needed to use multi-GPUs/TPU/fp16. Accelerate abstracts exactly and only the boilerplate code related to multi-GPUs/TPU/fp16 and leaves the rest of your code unchanged.

  • Homepage: https://github.com/huggingface/accelerate
  • Licenses: Apache-2.0
  • Latest release: 1.13.0 (published 6 months ago)
  • Last Synced: 2026-07-01T21:04:37.091Z (4 months ago)
  • Versions: 8
  • Dependent Packages: 1
  • Dependent Repositories: 0
  • Downloads: 6,589 Total
  • Rankings:
    • Stargazers count: 13.392%
    • Forks count: 18.904%
    • Dependent packages count: 20.494%
    • Average: 27.621%
    • Dependent repos count: 57.694%
anaconda.org: accelerate

Accelerate was created for PyTorch users who like to write the training loop of PyTorch models but are reluctant to write and maintain the boilerplate code needed to use multi-GPUs/TPU/fp16. Accelerate abstracts exactly and only the boilerplate code related to multi-GPUs/TPU/fp16 and leaves the rest of your code unchanged.

  • Homepage: https://github.com/huggingface/accelerate
  • Licenses: Apache-2.0
  • Latest release: 1.13.0 (published 6 months ago)
  • Last Synced: 2026-07-01T21:04:21.563Z (4 months ago)
  • Versions: 2
  • Dependent Packages: 0
  • Dependent Repositories: 0
  • Downloads: 70,770 Total
  • Rankings:
    • Downloads: 9.768%
    • Average: 28.067%
    • Dependent packages count: 35.608%
    • Dependent repos count: 38.826%
gentoo-portage: sci-ml/accelerate

Run your *raw* PyTorch training script on any kind of device

  • Homepage: https://github.com/huggingface/accelerate
  • Documentation: https://packages.gentoo.org/packages/sci-ml/accelerate
  • Licenses: Apache-2.0
  • Latest release: 1.14.0 (published 3 months ago)
  • Last Synced: 2026-08-12T01:03:48.609Z (3 months ago)
  • Versions: 2
  • Dependent Packages: 0
  • Dependent Repositories: 0
  • Rankings:
    • Dependent repos count: 0.0%
    • Dependent packages count: 0.0%
    • Average: 100%

Dependencies

.github/workflows/build-docker-images-release.yml actions
  • actions/checkout v3.1.0 composite
  • docker/build-push-action v4 composite
  • docker/login-action v2 composite
  • docker/setup-buildx-action v2 composite
.github/workflows/build_and_run_tests.yml actions
  • actions/checkout v3.1.0 composite
  • tj-actions/changed-files v41 composite
.github/workflows/build_docker_images.yml actions
  • docker/build-push-action v4 composite
  • docker/login-action v2 composite
  • docker/setup-buildx-action v2 composite
.github/workflows/build_documentation.yml actions
.github/workflows/build_pr_documentation.yml actions
.github/workflows/integration_tests.yml actions
  • actions/checkout v3.1.0 composite
  • actions/setup-python v3 composite
.github/workflows/nightly.yml actions
.github/workflows/quality.yml actions
  • actions/checkout v3.1.0 composite
  • actions/setup-python v3 composite
.github/workflows/run_merge_tests.yml actions
.github/workflows/self_hosted_integration_tests.yml actions
.github/workflows/stale.yml actions
  • actions/checkout v3.1.0 composite
  • actions/setup-python v3 composite
.github/workflows/test.yml actions
  • actions/cache v3 composite
  • actions/checkout v3.1.0 composite
  • actions/setup-python v3 composite
.github/workflows/upload_pr_documentation.yml actions
docker/accelerate-cpu/Dockerfile docker
  • python 3.8-slim build
docker/accelerate-gpu/Dockerfile docker
  • continuumio/miniconda3 latest build
  • nvidia/cuda 12.1.0-cudnn8-devel-ubuntu20.04 build
examples/requirements.txt pypi
  • accelerate *
  • datasets ==2.3.2
  • evaluate *
pyproject.toml pypi
setup.py pypi
  • numpy >=1.17
docker/accelerate-gpu-deepspeed/Dockerfile docker
  • continuumio/miniconda3 latest build
  • nvidia/cuda 12.1.0-cudnn8-devel-ubuntu20.04 build
examples/inference/pippy/requirements.txt pypi
  • accelerate *
  • pippy >=0.2.0