xref: /aosp_15_r20/external/executorch/README-wheel.md (revision 523fa7a60841cd1ecfb9cc4201f1ca8b03ed023a)
1**ExecuTorch** is a [PyTorch](https://pytorch.org/) platform that provides
2infrastructure to run PyTorch programs everywhere from AR/VR wearables to
3standard on-device iOS and Android mobile deployments. One of the main goals for
4ExecuTorch is to enable wider customization and deployment capabilities of the
5PyTorch programs.
6
7The `executorch` pip package is in alpha.
8* Supported python versions: 3.10, 3.11
9* Compatible systems: Linux x86_64, macOS aarch64
10
11The prebuilt `executorch.extension.pybindings.portable_lib` module included in
12this package provides a way to run ExecuTorch `.pte` files, with some
13restrictions:
14* Only [core ATen
15  operators](https://pytorch.org/executorch/stable/ir-ops-set-definition.html)
16  are linked into the prebuilt module
17* Only the [XNNPACK backend
18  delegate](https://pytorch.org/executorch/main/native-delegates-executorch-xnnpack-delegate.html)
19  is linked into the prebuilt module
20* [macOS only] [Core ML](https://pytorch.org/executorch/main/build-run-coreml.html) and [MPS](https://pytorch.org/executorch/main/build-run-mps.html) backend delegates are linked into the prebuilt module.
21
22Please visit the [ExecuTorch website](https://pytorch.org/executorch/) for
23tutorials and documentation. Here are some starting points:
24* [Getting
25  Started](https://pytorch.org/executorch/stable/getting-started-setup.html)
26  * Set up the ExecuTorch environment and run PyTorch models locally.
27* [Working with
28  local LLMs](https://pytorch.org/executorch/stable/llm/getting-started.html)
29  * Learn how to use ExecuTorch to export and accelerate a large-language model
30    from scratch.
31* [Exporting to
32  ExecuTorch](https://pytorch.org/executorch/main/tutorials/export-to-executorch-tutorial.html)
33  * Learn the fundamentals of exporting a PyTorch `nn.Module` to ExecuTorch, and
34    optimizing its performance using quantization and hardware delegation.
35* Running LLaMA on
36  [iOS](https://pytorch.org/executorch/stable/llm/llama-demo-ios.html) and
37  [Android](https://pytorch.org/executorch/stable/llm/llama-demo-android.html)
38  devices.
39  * Build and run LLaMA in a demo mobile app, and learn how to integrate models
40    with your own apps.
41