Torch-TensorRT

license

BSD-3-Clause(OSI)

source

public

core features withheld

no

BSD-3-Clause license body confirmed. The repository is public and unarchived and builds the whole product, and the README describes no paid tier, enterprise edition or license-gated build beside it, so source is public and the core ungated.

• https://api.github.com/repos/pytorch/TensorRT recorded 2026-09-02

  Repo metadata for torch-tensorrt's canonical repository - private false, archived false, fork false.

• https://github.com/pytorch/TensorRT/blob/main/LICENSE recorded 2026-10-07

  LICENSE body is the verbatim three-clause BSD text under an NVIDIA copyright line ("Copyright (c) 2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.").

• https://github.com/pytorch/TensorRT/blob/main/README.md recorded 2026-10-07

  README describes Torch-TensorRT as bringing "the power of TensorRT to PyTorch" to "achieve the best inference performance for any PyTorch model on the NVIDIA platform", through torch.compile or an ahead-of-time export workflow; it names no paid tier, enterprise edition or license-gated build beside the published source.

Adoption is measured from PyPI downloads of the `torch-tensorrt` package. NVIDIA also ships Torch-TensorRT preinstalled inside its PyTorch NGC containers, a channel with no published download count, so the measured figure is a floor.

• https://pypi.org/pypi/torch-tensorrt/json recorded 2026-09-02

  PyPI metadata for torch-tensorrt 2.13.0 - project_urls Repository = https://github.com/pytorch/tensorrt.git, summary 'Torch-TensorRT is a package which allows users to automatically compile PyTorch and TorchScript modules to TensorRT'.

• https://pypistats.org/api/packages/torch-tensorrt/recent recorded 2026-09-02

  last_month downloads = 33,392 for torch-tensorrt

Torch-TensorRT compiles PyTorch, TorchScript and FX graphs through TensorRT's optimizing builder, narrowed to one framework and one vendor's hardware. That scope places it one step below Apache TVM's broader reach.

• https://github.com/pytorch/TensorRT/blob/main/README.md recorded 2026-10-07

  README describes Torch-TensorRT as bringing "the power of TensorRT to PyTorch" to "achieve the best inference performance for any PyTorch model on the NVIDIA platform", through torch.compile or an ahead-of-time export workflow; it names no paid tier, enterprise edition or license-gated build beside the published source.