Torch Packaging Dependencies
This repository does not contain a machine learning model.
Instead, it provides a packaged set of unmodified runtime binaries required for applications that depend on:
- PyTorch
- TorchVision
- TorchAudio
- CUDA extension libraries (e.g., torch_cuda.dll)
These files are redistributed here solely for the purpose of installer‑time dependency delivery, enabling lightweight application installers to download the required PyTorch stack on demand.
📦 Contents
This repository includes:
- Prebuilt PyTorch binaries
- Prebuilt TorchVision binaries
- Prebuilt TorchAudio binaries
- CUDA extension DLLs required by the above packages
- Corresponding upstream LICENSE and NOTICE files
All binaries are unmodified and provided exactly as distributed by the official PyTorch project.
🔍 Source of the Binaries
The dependencies originate from the official upstream repositories:
- PyTorch: https://github.com/pytorch/pytorch
- TorchVision: https://github.com/pytorch/vision
- TorchAudio: https://github.com/pytorch/audio
CUDA extension binaries are part of the official PyTorch distribution.
📜 Licensing
All included binaries are licensed under the BSD‑3‑Clause License, as provided by the PyTorch project.
This repository does not alter, relicense, or modify any of the included components.
The following files are included to comply with redistribution requirements:
LICENSE-pytorchLICENSE-torchvisionLICENSE-torchaudioNOTICE(if present in upstream distributions)
For full details, refer to the upstream license documents.
🧩 Purpose of This Repository
This repository exists to support:
- Application installers (e.g., Inno Setup)
- Automated deployment pipelines
- Lightweight distribution of applications that depend on PyTorch
By hosting these binaries separately, application installers can remain small while still providing the required runtime dependencies.
⚠️ No Warranty / No Endorsement
This repository is not affiliated with, endorsed by, or supported by the PyTorch team, Meta AI, NVIDIA, or Hugging Face.
All trademarks and copyrights belong to their respective owners.
📬 Contact
For questions about this packaging repository, please contact the maintainer:
RoboTraxxInc
For issues related to PyTorch itself, please refer to the official PyTorch GitHub issue tracker.