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

https://github.com/pytorch/serve

cpu deep-learning docker gpu kubernetes machine-learning metrics mlops optimization pytorch serving

Score: 29.16759674106925

Last synced: about 7 hours ago
JSON representation

Repository metadata:

Serve, optimize and scale PyTorch models in production


Owner metadata:


GitHub Events

Total
Last Year

Committers metadata

Last synced: about 2 months ago

Total Commits: 2,893
Total Committers: 213
Avg Commits per committer: 13.582
Development Distribution Score (DDS): 0.821

Commits in past year: 8
Committers in past year: 4
Avg Commits per committer in past year: 2.0
Development Distribution Score (DDS) in past year: 0.5

Name Email Commits
harshbafna h****a@p****n 519
lxning l****a@a****m 443
Mark Saroufim m****m@f****m 314
Ankith Gunapal a****l@i****u 155
Hamid Shojanazeri h****0@g****m 137
jagadeesh j****j@i****m 114
dhaniram kshirsagar d****r@p****m 105
Matthias Reso 1****o 86
Aaqib Ansari m****b@g****m 76
dependabot[bot] 4****] 65
Shrinath Suresh s****h@i****m 63
Naman Nandan n****5@g****m 55
udaij12 3****2 41
alvarobartt a****6@u****s 39
Nikhil Kulkarni n****k@a****m 39
dhanainme 6****e 35
Shivam Shriwas s****1@g****m 33
Mahesh Ambule a****h@g****m 29
Nikhil Kulkarni k****9@g****m 26
eslesar-aws e****r@a****m 22
Aaron Markham m****a@a****m 21
Deepak Babel d****l@p****m 19
min-jean-cho m****o@i****m 19
Manoj Rao m****o@a****m 15
Prashant Sail p****4@g****m 14
Brad Heintz b****z@f****m 14
min-jean-cho m****c@m****m 13
sachanub 1****b 12
sekyondaMeta 1****a 12
Abishek Chiffon a****n@g****m 12
and 183 more...

Issue and Pull Request metadata

Last synced: 4 months ago

Total issues: 472
Total pull requests: 987
Average time to close issues: 2 months
Average time to close pull requests: 20 days
Total issue authors: 226
Total pull request authors: 102
Average comments per issue: 2.21
Average comments per pull request: 0.89
Merged pull request: 641
Bot issues: 84
Bot pull requests: 192

Past year issues: 35
Past year pull requests: 101
Past year average time to close issues: 2 days
Past year average time to close pull requests: 27 days
Past year issue authors: 30
Past year pull request authors: 20
Past year average comments per issue: 0.46
Past year average comments per pull request: 0.52
Past year merged pull request: 21
Past year bot issues: 0
Past year bot pull requests: 53

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

Top Issue Authors

  • github-actions[bot] (82)
  • agunapal (34)
  • lxning (31)
  • mreso (11)
  • msaroufim (10)
  • pengxin233 (9)
  • namannandan (7)
  • feeeper (6)
  • rbavery (6)
  • jack-gits (6)
  • chauhang (4)
  • udaij12 (4)
  • hungtrieu07 (4)
  • sachanub (4)
  • hungtooc (3)

Top Pull Request Authors

  • dependabot[bot] (192)
  • agunapal (181)
  • mreso (104)
  • udaij12 (90)
  • namannandan (76)
  • lxning (59)
  • msaroufim (49)
  • sachanub (12)
  • jagadeeshi2i (11)
  • sekyondaMeta (10)
  • HamidShojanazeri (10)
  • harshita-meena (9)
  • ravi9 (8)
  • jakki-amd (6)
  • shrinath-suresh (6)

Top Issue Labels

  • triaged (94)
  • enhancement (46)
  • bug (26)
  • triaged_wait (23)
  • internal (21)
  • documentation (18)
  • docker (11)
  • dependencies (11)
  • security (10)
  • debugging (9)
  • support (9)
  • kubernetes (9)
  • help wanted (8)
  • metrics (7)
  • question (7)
  • m1 (6)
  • benchmark (6)
  • example (6)
  • kfserving (5)
  • p0 (5)
  • workflowx (5)
  • perf (5)
  • python (4)
  • good first issue (4)
  • p2 (4)
  • large-model (3)
  • windows (3)
  • grpc (3)
  • p1 (3)
  • java (3)

Top Pull Request Labels

  • dependencies (196)
  • python (120)
  • java (68)
  • enhancement (25)
  • documentation (24)
  • torch.compile (24)
  • c++ (18)
  • example (18)
  • bootcamp (13)
  • p0 (12)
  • github_actions (10)
  • module: rocm (9)
  • security (9)
  • bug (7)
  • kubernetes (6)
  • ci (5)
  • docker (4)
  • code-quality (4)
  • kfserving (3)
  • preprocessing (3)
  • easy (2)
  • optimization (2)
  • llm (2)
  • grpc (2)
  • p1 (2)
  • aws (1)
  • p2 (1)
  • metrics (1)
  • large-model (1)

Package metadata

pypi.org: torch-model-archiver

Torch Model Archiver is used for creating archives of trained neural net models that can be consumed by TorchServe inference

  • Homepage: https://github.com/pytorch/serve/blob/master/model-archiver
  • Documentation: https://torch-model-archiver.readthedocs.io/
  • Licenses: Apache License Version 2.0
  • Latest release: 0.12.0 (published over 1 year ago)
  • Last Synced: 2026-01-14T13:31:45.892Z (23 days ago)
  • Versions: 25
  • Dependent Packages: 7
  • Dependent Repositories: 251
  • Downloads: 436,424 Last month
  • Docker Downloads: 1,343,800
  • Rankings:
    • Downloads: 0.486%
    • Docker downloads count: 0.797%
    • Dependent repos count: 0.947%
    • Average: 1.109%
    • Stargazers count: 1.288%
    • Forks count: 1.524%
    • Dependent packages count: 1.611%
  • Maintainers (4)
pypi.org: torchserve

TorchServe is a tool for serving neural net models for inference

pypi.org: torch-workflow-archiver

Torch Workflow Archiver is used for creating archives of workflow designed using trained neural net models that can be consumed by TorchServe inference

  • Homepage: https://github.com/pytorch/serve/blob/master/workflow-archiver/
  • Documentation: https://torch-workflow-archiver.readthedocs.io/
  • Licenses: Apache License Version 2.0
  • Latest release: 0.2.15 (published over 1 year ago)
  • Last Synced: 2026-01-14T10:09:22.388Z (23 days ago)
  • Versions: 19
  • Dependent Packages: 1
  • Dependent Repositories: 24
  • Downloads: 4,861 Last month
  • Docker Downloads: 1,336,837
  • Rankings:
    • Docker downloads count: 0.798%
    • Stargazers count: 1.287%
    • Forks count: 1.524%
    • Downloads: 2.398%
    • Average: 2.727%
    • Dependent repos count: 2.98%
    • Dependent packages count: 7.377%
  • Maintainers (4)
pypi.org: torch-workflow-archiver-nightly

Torch Workflow Archiver is used for creating archives of workflow designed using trained neural net models that can be consumed by TorchServe inference

  • Homepage: https://github.com/pytorch/serve/blob/master/workflow-archiver/
  • Documentation: https://torch-workflow-archiver-nightly.readthedocs.io/
  • Licenses: Apache License Version 2.0
  • Latest release: 2025.1.13 (published about 1 year ago)
  • Last Synced: 2025-01-14T10:06:54.246Z (about 1 year ago)
  • Versions: 1,043
  • Dependent Packages: 0
  • Dependent Repositories: 1
  • Downloads: 8,786 Last month
  • Rankings:
    • Stargazers count: 1.297%
    • Forks count: 1.549%
    • Downloads: 3.484%
    • Average: 7.608%
    • Dependent packages count: 10.095%
    • Dependent repos count: 21.615%
  • Maintainers (2)
pypi.org: torchserve-nightly

TorchServe is a tool for serving neural net models for inference

  • Homepage: https://github.com/pytorch/serve.git
  • Documentation: https://torchserve-nightly.readthedocs.io/
  • Licenses: Apache License Version 2.0
  • Latest release: 2025.1.13 (published about 1 year ago)
  • Last Synced: 2025-01-14T10:20:51.792Z (about 1 year ago)
  • Versions: 1,056
  • Dependent Packages: 0
  • Dependent Repositories: 1
  • Downloads: 9,341 Last month
  • Rankings:
    • Stargazers count: 1.297%
    • Forks count: 1.549%
    • Downloads: 3.646%
    • Average: 7.641%
    • Dependent packages count: 10.095%
    • Dependent repos count: 21.615%
  • Maintainers (1)
pypi.org: torch-model-archiver-nightly

Torch Model Archiver is used for creating archives of trained neural net models that can be consumed by TorchServe inference

  • Homepage: https://github.com/pytorch/serve/blob/master/model-archiver
  • Documentation: https://torch-model-archiver-nightly.readthedocs.io/
  • Licenses: Apache License Version 2.0
  • Latest release: 2025.1.13 (published about 1 year ago)
  • Last Synced: 2025-01-14T10:21:02.222Z (about 1 year ago)
  • Versions: 1,044
  • Dependent Packages: 0
  • Dependent Repositories: 1
  • Downloads: 14,371 Last month
  • Rankings:
    • Stargazers count: 1.297%
    • Forks count: 1.549%
    • Downloads: 3.661%
    • Average: 7.644%
    • Dependent packages count: 10.095%
    • Dependent repos count: 21.615%
  • Maintainers (2)
proxy.golang.org: github.com/pytorch/serve

  • Homepage:
  • Documentation: https://pkg.go.dev/github.com/pytorch/serve#section-documentation
  • Licenses: apache-2.0
  • Latest release: v0.12.0 (published over 1 year ago)
  • Last Synced: 2026-01-04T02:45:20.614Z (about 1 month ago)
  • Versions: 24
  • Dependent Packages: 0
  • Dependent Repositories: 0
  • Rankings:
    • Dependent packages count: 6.999%
    • Average: 8.173%
    • Dependent repos count: 9.346%
pypi.org: torch-model-archiver-ag

Torch Model Archiver is used for creating archives of trained neural net models that can be consumed by TorchServe inference

  • Homepage: https://github.com/pytorch/serve/model-archiver/
  • Documentation: https://torch-model-archiver-ag.readthedocs.io/
  • Licenses: Apache License Version 2.0
  • Latest release: 0.6.0b20220624 (published over 3 years ago)
  • Last Synced: 2026-01-09T14:37:54.553Z (28 days ago)
  • Versions: 1
  • Dependent Packages: 0
  • Dependent Repositories: 0
  • Downloads: 8 Last month
  • Rankings:
    • Stargazers count: 1.305%
    • Forks count: 1.581%
    • Dependent packages count: 6.633%
    • Average: 17.997%
    • Dependent repos count: 30.611%
    • Downloads: 49.858%
  • Maintainers (1)
pypi.org: torchserve-ag

TorchServe is a tool for serving neural net models for inference

  • Homepage: https://github.com/pytorch/serve.git
  • Documentation: https://torchserve-ag.readthedocs.io/
  • Licenses: Apache License Version 2.0
  • Latest release: 0.8.2b20230918 (published over 2 years ago)
  • Last Synced: 2026-01-12T08:58:08.551Z (25 days ago)
  • Versions: 2
  • Dependent Packages: 0
  • Dependent Repositories: 0
  • Downloads: 14 Last month
  • Rankings:
    • Stargazers count: 1.305%
    • Forks count: 1.581%
    • Dependent packages count: 6.633%
    • Average: 19.25%
    • Dependent repos count: 30.611%
    • Downloads: 56.12%
  • Maintainers (1)
pypi.org: torch-workflow-archiver-ag

Torch Workflow Archiver is used for creating archives of workflow designed using trained neural net models that can be consumed by TorchServe inference

  • Homepage: https://github.com/pytorch/serve/workflow-archiver/
  • Documentation: https://torch-workflow-archiver-ag.readthedocs.io/
  • Licenses: Apache License Version 2.0
  • Latest release: 0.2.4b20220624 (published over 3 years ago)
  • Last Synced: 2026-01-12T01:09:56.007Z (25 days ago)
  • Versions: 1
  • Dependent Packages: 0
  • Dependent Repositories: 0
  • Downloads: 13 Last month
  • Rankings:
    • Stargazers count: 1.305%
    • Forks count: 1.581%
    • Dependent packages count: 6.633%
    • Average: 20.727%
    • Dependent repos count: 30.611%
    • Downloads: 63.504%
  • Maintainers (1)

Dependencies

.github/workflows/ci_cpu.yml actions
  • actions/checkout v3 composite
  • actions/setup-java v3 composite
  • actions/setup-python v4 composite
.github/workflows/ci_gpu.yml actions
  • actions/checkout v3 composite
  • actions/setup-java v3 composite
  • actions/setup-python v4 composite
.github/workflows/doc-automation.yml actions
  • JamesIves/github-pages-deploy-action releases/v3 composite
  • actions/checkout v3 composite
  • actions/setup-python v4 composite
  • dacbd/create-issue-action v1 composite
.github/workflows/docker-nightly-build.yml actions
  • actions/checkout v3 composite
  • actions/setup-python v4 composite
  • dacbd/create-issue-action v1 composite
.github/workflows/lint.yml actions
  • actions/checkout v3 composite
  • actions/setup-python v4 composite
  • gaurav-nelson/github-action-markdown-link-check 1.0.13 composite
  • tj-actions/changed-files v29.0.4 composite
.github/workflows/official_release.yml actions
  • actions/checkout v3 composite
  • actions/setup-java v2 composite
  • s-weigand/setup-conda v1 composite
.github/workflows/regression_tests_cpu.yml actions
  • actions/checkout v3 composite
  • actions/setup-java v3 composite
  • actions/setup-python v3 composite
.github/workflows/regression_tests_gpu.yml actions
  • actions/checkout v3 composite
  • actions/setup-java v3 composite
  • actions/setup-python v3 composite
.github/workflows/torchserve-nightly-build.yml actions
  • actions/checkout v3 composite
  • actions/setup-java v2 composite
  • dacbd/create-issue-action v1 composite
  • s-weigand/setup-conda v1 composite
docker/Dockerfile docker
  • ${BASE_IMAGE} latest build
benchmarks/requirements-ab.txt pypi
  • Click *
  • click-config-file *
  • matplotlib *
  • mdutils *
  • pandas *
  • pyyaml *
  • requests *
  • ruamel.yaml *
ci/launcher/requirements.txt pypi
  • boto3 *
  • fabric2 *
  • retrying *
docs/sphinx/requirements.txt pypi
  • katex *
  • markdown <3.4.0
  • recommonmark ==0.6.0
  • sphinx ==4.0.3
  • sphinx-markdown-tables ==0.0.15
  • sphinxcontrib.katex *
  • torch *
  • torch-model-archiver *
  • torch-workflow-archiver *
  • torchserve *
examples/diffusers/requirements.txt pypi
  • diffusers ==0.6.0
  • numpy *
  • transformers *
.github/workflows/benchmark_nightly.yml actions
  • actions/checkout v3 composite
  • actions/setup-java v3 composite
  • actions/setup-python v4 composite
  • actions/upload-artifact v2 composite
  • dacbd/create-issue-action v1 composite
  • dawidd6/action-download-artifact v2 composite
.github/workflows/codeql.yml actions
  • actions/checkout v3 composite
  • actions/setup-java v3 composite
  • actions/setup-python v4 composite
  • github/codeql-action/analyze v2 composite
  • github/codeql-action/init v2 composite
.github/workflows/docker-ci.yaml actions
  • actions/checkout v3 composite
.github/workflows/regression_tests_cpu_binaries.yml actions
  • actions/checkout v3 composite
  • actions/setup-java v3 composite
  • s-weigand/setup-conda v1 composite
.github/workflows/regression_tests_docker.yml actions
  • actions/checkout v3 composite
  • dacbd/create-issue-action v1 composite
.github/workflows/regression_tests_gpu_binaries.yml actions
  • actions/checkout v3 composite
  • actions/setup-java v3 composite
  • actions/setup-python v3 composite
  • conda-incubator/setup-miniconda v2 composite
  • s-weigand/setup-conda v1 composite
examples/Huggingface_Transformers/requirements.txt pypi
  • optimum *
  • transformers *
examples/asr_rnnt_emformer/requirements.txt pypi
  • sentencepiece *
  • torchaudio *
.github/workflows/official_release_docker.yml actions
  • actions/checkout v3 composite
  • actions/setup-python v4 composite
.github/workflows/benchmark_torch_compile_nightly.yml actions
  • actions/checkout v3 composite
  • actions/setup-java v3 composite
  • actions/setup-python v4 composite
.github/workflows/kserve_cpu_tests.yml actions
  • actions/checkout v4 composite
  • actions/checkout v3 composite
  • actions/setup-python v4 composite
kubernetes/kserve/Dockerfile docker
  • ${BASE_IMAGE} latest build
frontend/archive/build.gradle maven
  • com.google.code.gson:gson ${gson_version} api
  • commons-io:commons-io 2.15.1 api
  • org.apache.logging.log4j:log4j-slf4j-impl ${slf4j_log4j_version} api
  • org.slf4j:slf4j-api ${slf4j_api_version} api
  • org.apache.commons:commons-compress 1.23.0 implementation
  • org.yaml:snakeyaml ${snakeyaml_version} implementation
  • commons-cli:commons-cli ${commons_cli_version} testImplementation
  • org.testng:testng ${testng_version} testImplementation
frontend/build.gradle maven
frontend/server/build.gradle maven
  • com.lmax:disruptor ${lmax_disruptor_version} implementation
  • commons-cli:commons-cli ${commons_cli_version} implementation
  • io.netty:netty-all ${netty_version} implementation
  • io.prometheus:simpleclient ${prometheus_version} implementation
  • io.prometheus:simpleclient_servlet ${prometheus_version} implementation
  • org.apache.logging.log4j:log4j-core ${slf4j_log4j_version} implementation
  • org.pytorch:torchserve-plugins-sdk ${torchserve_sdk_version} implementation
  • org.yaml:snakeyaml ${snakeyaml_version} implementation
  • org.testng:testng ${testng_version} testImplementation
plugins/DDBEndPoint/build.gradle maven
  • com.google.code.gson:gson ${gson_version} implementation
  • org.codehaus.groovy:groovy-all 2.3.11 implementation
  • org.pytorch:torchserve-plugins-sdk ${torchserve_sdk_version} implementation
  • software.amazon.awssdk:dynamodb-enhanced * implementation
  • com.amazonaws:DynamoDBLocal 2.1.0 testImplementation
  • org.testng:testng ${testng_version} testImplementation
plugins/build.gradle maven
plugins/endpoints/build.gradle maven
  • com.google.code.gson:gson ${gson_version} implementation
  • org.pytorch:torchserve-plugins-sdk ${torchserve_sdk_version} implementation
serving-sdk/pom.xml maven
  • com.google.code.gson:gson 2.10.1
  • junit:junit 4.13.1 test
  • org.mockito:mockito-core 4.6.1 test
examples/instruction_embedding/requirements.txt pypi
  • InstructorEmbedding *
examples/large_models/Huggingface_accelerate/requirements.txt pypi
  • accelerate *
  • transformers *
examples/large_models/Huggingface_pippy/requirements.txt pypi
  • torchpippy *
  • transformers *
examples/large_models/deepspeed/requirements.txt pypi
  • deepspeed >=0.9.4
  • transformers >=4.30.1
examples/large_models/deepspeed_mii/requirements.txt pypi
  • deepspeed-mii *
  • numpy *
  • transformers *
examples/large_models/inferentia2/opt/requirements.txt pypi
  • torch-neuronx *
  • transformers *
  • transformers-neuronx *
examples/nvidia_dali/requirements.txt pypi
  • nvidia-dali-cuda110 >=1.27.0
examples/object_detector/yolo/yolov8/requirements.txt pypi
  • ultralytics >=8.0.144
examples/stateful/requirements.txt pypi
  • lru-dict *
examples/torch_tensorrt/requirements.txt pypi
  • tensorrt ==8.5.3.1
  • torch-tensorrt ==1.4.0
kubernetes/kserve/image_transformer/setup.py pypi
  • argparse >=1.4.0
  • joblib >=0.13.2
  • kserve >=0.9.0
  • kubernetes >=9.0.0
  • numpy >=1.16.3
  • pandas >=0.24.2
  • pillow >=9.0.1
  • requests >=2.22.0
  • torchvision >=0.6.0
kubernetes/kserve/requirements.txt pypi
  • captum *
  • grpcio *
  • grpcio-tools *
  • kserve >=0.11.0
  • protobuf *
  • transformers *
model-archiver/setup.py pypi
requirements/common.txt pypi
  • captum ==0.6.0
  • packaging ==23.2
  • psutil ==5.9.5
  • pynvml ==11.5.0
  • pyyaml ==6.0
  • requests ==2.31.0
requirements/common_gpu.txt pypi
  • nvgpu *
  • nvgpu ==0.10.0
requirements/developer.txt pypi
  • googleapis-common-protos * development
  • grpcio ==1.54.2 development
  • grpcio-tools ==1.54.2 development
  • intel_extension_for_pytorch ==2.1.0 development
  • mock ==5.1.0 development
  • mypy ==1.3.0 development
  • onnx ==1.14.1 development
  • onnxruntime ==1.15.0 development
  • pre-commit ==3.3.2 development
  • protobuf ==4.25.1 development
  • pygit2 ==1.13.1 development
  • pylint ==2.6.0 development
  • pyspelling ==2.8.2 development
  • pytest ==7.3.1 development
  • pytest-cov ==4.1.0 development
  • pytest-mock ==3.12.0 development
  • torchpippy ==0.1.1 development
  • transformers >=4.34.0 development
  • twine ==4.0.2 development
requirements/neuronx.txt pypi
  • grpcio-tools ==1.48.2
  • neuronx-cc *
  • numpy ==1.22.4
  • protobuf ==4.25.1
  • torch-neuronx *
  • transformers-neuronx *
requirements/torch_common.txt pypi
  • cython ==3.0.5
  • numpy ==1.24.3
  • pillow ==10.0.1
  • wheel ==0.40.0
requirements/torch_cu101_linux.txt pypi
  • torch ==1.8.1
  • torchaudio ==0.8.1
  • torchtext ==0.9.1
  • torchvision ==0.9.1
requirements/torch_cu101_windows.txt pypi
  • torch ==1.8.1
  • torchaudio ==0.8.1
  • torchtext ==0.9.1
  • torchvision ==0.9.1
requirements/torch_cu102_linux.txt pypi
  • torch ==1.12.0
  • torchaudio ==0.12.0
  • torchtext ==0.13.0
  • torchvision ==0.13.0
requirements/torch_cu102_windows.txt pypi
  • torch ==1.10.0
  • torchaudio ==0.10.0
  • torchtext ==0.11.0
  • torchvision ==0.11.1
requirements/torch_cu111_linux.txt pypi
  • torch ==1.9.0
  • torchaudio ==0.9.0
  • torchtext ==0.10.0
  • torchvision ==0.10.0
requirements/torch_cu111_windows.txt pypi
  • torch ==1.9.0
  • torchaudio ==0.9.0
  • torchtext ==0.10.0
  • torchvision ==0.10.0
requirements/torch_cu113_linux.txt pypi
  • torch ==1.12.0
  • torchaudio ==0.12.0
  • torchtext ==0.13.0
  • torchvision ==0.13.0
requirements/torch_cu113_windows.txt pypi
  • torch ==1.12.0
  • torchaudio ==0.12.0
  • torchtext ==0.13.0
  • torchvision ==0.13.0
requirements/torch_cu116_linux.txt pypi
  • torch ==1.13.1
  • torchaudio ==0.13.1
  • torchtext ==0.14.1
  • torchvision ==0.14.1
requirements/torch_cu116_windows.txt pypi
  • torch ==1.13.1
  • torchaudio ==0.13.1
  • torchtext ==0.14.1
  • torchvision ==0.14.1
requirements/torch_cu117_linux.txt pypi
  • torch ==2.0.1
  • torchaudio ==2.0.2
  • torchtext ==0.15.2
  • torchvision ==0.15.2
requirements/torch_cu117_windows.txt pypi
  • torch ==2.0.1
  • torchaudio ==2.0.2
  • torchtext ==0.15.2
  • torchvision ==0.15.2
requirements/torch_cu118_linux.txt pypi
  • torch ==2.1.0
  • torchaudio ==2.1.0
  • torchtext ==0.16.0
  • torchvision ==0.16.0
requirements/torch_cu118_windows.txt pypi
  • torch ==2.1.0
  • torchaudio ==2.1.0
  • torchtext ==0.16.0
  • torchvision ==0.16.0
requirements/torch_cu121_linux.txt pypi
  • torch ==2.1.0
  • torchaudio ==2.1.0
  • torchtext ==0.16.0
  • torchvision ==0.16.0
requirements/torch_cu121_windows.txt pypi
  • torch ==2.1.0
  • torchaudio ==2.1.0
  • torchtext ==0.16.0
  • torchvision ==0.16.0
requirements/torch_cu92.txt pypi
  • torch ==1.8.1
  • torchaudio ==0.8.1
  • torchtext ==0.9.1
  • torchvision ==0.9.1
requirements/torch_cu92_linux.txt pypi
  • torch ==1.8.1
  • torchaudio ==0.8.1
  • torchtext ==0.9.1
  • torchvision ==0.9.1
requirements/torch_darwin.txt pypi
  • torch ==2.1.0
  • torchaudio ==2.1.0
  • torchtext ==0.16.0
  • torchvision ==0.16.0
requirements/torch_linux.txt pypi
  • torch ==2.1.0
  • torchaudio ==2.1.0
  • torchtext ==0.16.0
  • torchvision ==0.16.0
requirements/torch_neuronx_linux.txt pypi
  • torch ==1.13.1
  • torchaudio ==0.13.1
  • torchdata ==0.7.0
  • torchtext ==0.14.1
  • torchvision ==0.14.1
requirements/torch_windows.txt pypi
  • torch ==2.1.0
  • torchaudio ==2.1.0
  • torchtext ==0.16.0
  • torchvision ==0.16.0
setup.py pypi
ts/handler_utils/preprocess/built-in/requirements.txt pypi
  • nvidia-dali-cuda110 >=1.27.0
workflow-archiver/setup.py pypi
examples/large_models/deepspeed_mii/LLM/requirements.txt pypi
  • deepspeed-mii *
kubernetes/kserve/examples/gpt_fast/Dockerfile docker
  • pytorch/torchserve-kfs-nightly latest-gpu build
.github/workflows/ci-cpu-cpp.yml actions
  • actions/checkout v2 composite
test/pytest/test_data/custom_dependencies/requirements.txt pypi
  • matplotlib * test
.github/workflows/kubernetes_tests.yml actions
  • actions/checkout v3 composite
  • actions/setup-python v5 composite
kubernetes/tests/docker/Dockerfile docker
  • pytorch/torchserve-nightly latest-gpu build
examples/large_models/Huggingface_accelerate/llama/requirements.txt pypi
  • accelerate *
  • bitsandbytes *
  • mpi4py *
  • optimum *
  • scipy *
  • transformers >=4.38.0
examples/xgboost_classfication/requirements.txt pypi
  • scikit-learn *
  • xgboost *
examples/LLM/llama/chat_app/docker/Dockerfile docker
  • $BASE_IMAGE latest build
examples/LLM/llama/chat_app/docker/requirements.txt pypi
  • llama-cpp-python *
  • requests_futures *
  • streamlit >=1.26.0
  • transformers *
examples/LLM/llama/chat_app/requirements.txt pypi
  • streamlit >=1.26.0
examples/large_models/inferentia2/llama/requirements.txt pypi
  • sentencepiece *
  • tokenizers ==0.19.1
  • transformers ==4.40.0
examples/large_models/vllm/requirements.txt pypi
  • vllm *
examples/pt2/torch_compile_openvino/requirements.txt pypi
  • openvino >=2024.1.0
examples/pt2/torch_compile_openvino/stable_diffusion/requirements.txt pypi
  • diffusers >=0.27.2
  • openvino >=2024.1.0
  • torch >=2.1.0
examples/text_to_speech_synthesizer/WaveGlow/requirements.txt pypi
  • inflect *
  • librosa *
  • numpy *
  • scipy *
  • unidecode *
requirements/torch_linux_aarch64.txt pypi
  • torch ==2.2.1
  • torchaudio ==2.2.1
  • torchvision ==0.17.1
examples/speech2text_wav2vec2/environment.yml conda
  • pip
  • python >=3.10
  • pytorch
  • torchaudio
  • transformers
examples/large_models/ipex_llm_int8/requirements.txt pypi
  • accelerate *
  • datasets ==2.18.0
  • intel_extension_for_pytorch ==2.4.0
  • torch ==2.4.0.dev20240328
  • torchserve ==0.10.0
  • transformers ==4.38.2