uv/docs/guides/integration/pytorch.md

title	description
Using uv with PyTorch	A guide to using uv with PyTorch, including installing PyTorch, configuring per-platform and per-accelerator builds, and more.

Using uv with PyTorch

The PyTorch ecosystem is a popular choice for deep learning research and development. You can use uv to manage PyTorch projects and PyTorch dependencies across different Python versions and environments, even controlling for the choice of accelerator (e.g., CPU-only vs. CUDA).

!!! note

Some of the features outlined in this guide require uv version 0.5.3 or later. We recommend upgrading prior to configuring PyTorch.

Installing PyTorch

From a packaging perspective, PyTorch has a few uncommon characteristics:

• Many PyTorch wheels are hosted on a dedicated index, rather than the Python Package Index (PyPI). As such, installing PyTorch often requires configuring a project to use the PyTorch index.
• PyTorch produces distinct builds for each accelerator (e.g., CPU-only, CUDA). Since there's no standardized mechanism for specifying these accelerators when publishing or installing, PyTorch encodes them in the local version specifier. As such, PyTorch versions will often look like 2.5.1+cpu, 2.5.1+cu121, etc.
• Builds for different accelerators are published to different indexes. For example, the +cpu builds are published on https://download.pytorch.org/whl/cpu, while the +cu121 builds are published on https://download.pytorch.org/whl/cu121.

As such, the necessary packaging configuration will vary depending on both the platforms you need to support and the accelerators you want to enable.

To start, consider the following (default) configuration, which would be generated by running uv init --python 3.12 followed by uv add torch torchvision.

In this case, PyTorch would be installed from PyPI, which hosts CPU-only wheels for Windows and macOS, and GPU-accelerated wheels on Linux (targeting CUDA 12.6):

[project]
name = "project"
version = "0.1.0"
requires-python = ">=3.12"
dependencies = [
  "torch>=2.7.0",
  "torchvision>=0.22.0",
]

!!! tip "Supported Python versions"

At time of writing, PyTorch does not yet publish wheels for Python 3.14; as such projects with
`requires-python = ">=3.14"` may fail to resolve. See the
[compatibility matrix](https://github.com/pytorch/pytorch/blob/main/RELEASE.md#release-compatibility-matrix).

This is a valid configuration for projects that want to use CPU builds on Windows and macOS, and CUDA-enabled builds on Linux. However, if you need to support different platforms or accelerators, you'll need to configure the project accordingly.

Using a PyTorch index

In some cases, you may want to use a specific PyTorch variant across all platforms. For example, you may want to use the CPU-only builds on Linux too.

In such cases, the first step is to add the relevant PyTorch index to your pyproject.toml:

=== "CPU-only"

```toml
[[tool.uv.index]]
name = "pytorch-cpu"
url = "https://download.pytorch.org/whl/cpu"
explicit = true
```

=== "CUDA 11.8"

```toml
[[tool.uv.index]]
name = "pytorch-cu118"
url = "https://download.pytorch.org/whl/cu118"
explicit = true
```

=== "CUDA 12.6"

```toml
[[tool.uv.index]]
name = "pytorch-cu126"
url = "https://download.pytorch.org/whl/cu126"
explicit = true
```

=== "CUDA 12.8"

```toml
[[tool.uv.index]]
name = "pytorch-cu128"
url = "https://download.pytorch.org/whl/cu128"
explicit = true
```

=== "ROCm6"

```toml
[[tool.uv.index]]
name = "pytorch-rocm"
url = "https://download.pytorch.org/whl/rocm6.3"
explicit = true
```

=== "Intel GPUs"

```toml
[[tool.uv.index]]
name = "pytorch-xpu"
url = "https://download.pytorch.org/whl/xpu"
explicit = true
```

We recommend the use of explicit = true to ensure that the index is only used for torch, torchvision, and other PyTorch-related packages, as opposed to generic dependencies like jinja2, which should continue to be sourced from the default index (PyPI).

Next, update the pyproject.toml to point torch and torchvision to the desired index:

=== "CPU-only"

```toml
[tool.uv.sources]
torch = [
  { index = "pytorch-cpu" },
]
torchvision = [
  { index = "pytorch-cpu" },
]
```

=== "CUDA 11.8"

PyTorch doesn't publish CUDA builds for macOS. As such, we gate on `sys_platform` to instruct uv to use
the PyTorch index on Linux and Windows, but fall back to PyPI on macOS:

```toml
[tool.uv.sources]
torch = [
  { index = "pytorch-cu118", marker = "sys_platform == 'linux' or sys_platform == 'win32'" },
]
torchvision = [
  { index = "pytorch-cu118", marker = "sys_platform == 'linux' or sys_platform == 'win32'" },
]
```

=== "CUDA 12.6"

PyTorch doesn't publish CUDA builds for macOS. As such, we gate on `sys_platform` to instruct uv to limit
the PyTorch index to Linux and Windows, falling back to PyPI on macOS:

```toml
[tool.uv.sources]
torch = [
  { index = "pytorch-cu126", marker = "sys_platform == 'linux' or sys_platform == 'win32'" },
]
torchvision = [
  { index = "pytorch-cu126", marker = "sys_platform == 'linux' or sys_platform == 'win32'" },
]
```

=== "CUDA 12.8"

PyTorch doesn't publish CUDA builds for macOS. As such, we gate on `sys_platform` to instruct uv to limit
the PyTorch index to Linux and Windows, falling back to PyPI on macOS:

```toml
[tool.uv.sources]
torch = [
  { index = "pytorch-cu128", marker = "sys_platform == 'linux' or sys_platform == 'win32'" },
]
torchvision = [
  { index = "pytorch-cu128", marker = "sys_platform == 'linux' or sys_platform == 'win32'" },
]
```

=== "ROCm6"

PyTorch doesn't publish ROCm6 builds for macOS or Windows. As such, we gate on `sys_platform` to instruct uv
to limit the PyTorch index to Linux, falling back to PyPI on macOS and Windows:

```toml
[tool.uv.sources]
torch = [
  { index = "pytorch-rocm", marker = "sys_platform == 'linux'" },
]
torchvision = [
  { index = "pytorch-rocm", marker = "sys_platform == 'linux'" },
]
# ROCm6 support relies on `pytorch-triton-rocm`, which should also be installed from the PyTorch index
# (and included in `project.dependencies`).
pytorch-triton-rocm = [
  { index = "pytorch-rocm", marker = "sys_platform == 'linux'" },
]
```

=== "Intel GPUs"

PyTorch doesn't publish Intel GPU builds for macOS. As such, we gate on `sys_platform` to instruct uv to limit
the PyTorch index to Linux and Windows, falling back to PyPI on macOS:

```toml
[tool.uv.sources]
torch = [
  { index = "pytorch-xpu", marker = "sys_platform == 'linux' or sys_platform == 'win32'" },
]
torchvision = [
  { index = "pytorch-xpu", marker = "sys_platform == 'linux' or sys_platform == 'win32'" },
]
# Intel GPU support relies on `pytorch-triton-xpu`, which should also be installed from the PyTorch index
# (and included in `project.dependencies`).
pytorch-triton-xpu = [
  { index = "pytorch-xpu", marker = "sys_platform == 'linux' or sys_platform == 'win32'" },
]
```

As a complete example, the following project would use PyTorch's CPU-only builds on all platforms:

[project]
name = "project"
version = "0.1.0"
requires-python = ">=3.12.0"
dependencies = [
  "torch>=2.7.0",
  "torchvision>=0.22.0",
]

[tool.uv.sources]
torch = [
    { index = "pytorch-cpu" },
]
torchvision = [
    { index = "pytorch-cpu" },
]

[[tool.uv.index]]
name = "pytorch-cpu"
url = "https://download.pytorch.org/whl/cpu"
explicit = true

Configuring accelerators with environment markers

In some cases, you may want to use CPU-only builds in one environment (e.g., macOS and Windows), and CUDA-enabled builds in another (e.g., Linux).

With tool.uv.sources, you can use environment markers to specify the desired index for each platform. For example, the following configuration would use PyTorch's CUDA-enabled builds on Linux, and CPU-only builds on all other platforms (e.g., macOS and Windows):

[project]
name = "project"
version = "0.1.0"
requires-python = ">=3.12.0"
dependencies = [
  "torch>=2.7.0",
  "torchvision>=0.22.0",
]

[tool.uv.sources]
torch = [
  { index = "pytorch-cpu", marker = "sys_platform != 'linux'" },
  { index = "pytorch-cu128", marker = "sys_platform == 'linux'" },
]
torchvision = [
  { index = "pytorch-cpu", marker = "sys_platform != 'linux'" },
  { index = "pytorch-cu128", marker = "sys_platform == 'linux'" },
]

[[tool.uv.index]]
name = "pytorch-cpu"
url = "https://download.pytorch.org/whl/cpu"
explicit = true

[[tool.uv.index]]
name = "pytorch-cu128"
url = "https://download.pytorch.org/whl/cu128"
explicit = true

Similarly, the following configuration would use PyTorch's AMD GPU builds on Linux, and CPU-only builds on Windows and macOS (by way of falling back to PyPI):

[project]
name = "project"
version = "0.1.0"
requires-python = ">=3.12.0"
dependencies = [
  "torch>=2.7.0",
  "torchvision>=0.22.0",
  "pytorch-triton-rocm>=3.3.0 ; sys_platform == 'linux'",
]

[tool.uv.sources]
torch = [
  { index = "pytorch-rocm", marker = "sys_platform == 'linux'" },
]
torchvision = [
  { index = "pytorch-rocm", marker = "sys_platform == 'linux'" },
]
pytorch-triton-rocm = [
  { index = "pytorch-rocm", marker = "sys_platform == 'linux'" },
]

[[tool.uv.index]]
name = "pytorch-rocm"
url = "https://download.pytorch.org/whl/rocm6.3"
explicit = true

Or, for Intel GPU builds:

[project]
name = "project"
version = "0.1.0"
requires-python = ">=3.12.0"
dependencies = [
  "torch>=2.7.0",
  "torchvision>=0.22.0",
  "pytorch-triton-xpu>=3.3.0 ; sys_platform == 'win32' or sys_platform == 'linux'",
]

[tool.uv.sources]
torch = [
  { index = "pytorch-xpu", marker = "sys_platform == 'win32' or sys_platform == 'linux'" },
]
torchvision = [
  { index = "pytorch-xpu", marker = "sys_platform == 'win32' or sys_platform == 'linux'" },
]
pytorch-triton-xpu = [
  { index = "pytorch-xpu", marker = "sys_platform == 'win32' or sys_platform == 'linux'" },
]

[[tool.uv.index]]
name = "pytorch-xpu"
url = "https://download.pytorch.org/whl/xpu"
explicit = true

Configuring accelerators with optional dependencies

In some cases, you may want to use CPU-only builds in some cases, but CUDA-enabled builds in others, with the choice toggled by a user-provided extra (e.g., uv sync --extra cpu vs. uv sync --extra cu128).

With tool.uv.sources, you can use extra markers to specify the desired index for each enabled extra. For example, the following configuration would use PyTorch's CPU-only for uv sync --extra cpu and CUDA-enabled builds for uv sync --extra cu128:

[project]
name = "project"
version = "0.1.0"
requires-python = ">=3.12.0"
dependencies = []

[project.optional-dependencies]
cpu = [
  "torch>=2.7.0",
  "torchvision>=0.22.0",
]
cu128 = [
  "torch>=2.7.0",
  "torchvision>=0.22.0",
]

[tool.uv]
conflicts = [
  [
    { extra = "cpu" },
    { extra = "cu128" },
  ],
]

[tool.uv.sources]
torch = [
  { index = "pytorch-cpu", extra = "cpu" },
  { index = "pytorch-cu128", extra = "cu128" },
]
torchvision = [
  { index = "pytorch-cpu", extra = "cpu" },
  { index = "pytorch-cu128", extra = "cu128" },
]

[[tool.uv.index]]
name = "pytorch-cpu"
url = "https://download.pytorch.org/whl/cpu"
explicit = true

[[tool.uv.index]]
name = "pytorch-cu128"
url = "https://download.pytorch.org/whl/cu128"
explicit = true

!!! note

Since GPU-accelerated builds aren't available on macOS, the above configuration will fail to install
on macOS when the `cu128` extra is enabled.

The uv pip interface

While the above examples are focused on uv's project interface (uv lock, uv sync, uv run, etc.), PyTorch can also be installed via the uv pip interface.

PyTorch itself offers a dedicated interface to determine the appropriate pip command to run for a given target configuration. For example, you can install stable, CPU-only PyTorch on Linux with:

$ pip3 install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cpu

To use the same workflow with uv, replace pip3 with uv pip:

$ uv pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cpu

Automatic backend selection

uv supports automatic selection of the appropriate PyTorch index via the --torch-backend=auto command-line argument (or the UV_TORCH_BACKEND=auto environment variable), as in:

$ # With a command-line argument.
$ uv pip install torch --torch-backend=auto

$ # With an environment variable.
$ UV_TORCH_BACKEND=auto uv pip install torch

When enabled, uv will query for the installed CUDA driver and AMD GPU versions then use the most-compatible PyTorch index for all relevant packages (e.g., torch, torchvision, etc.). If no such GPU is found, uv will fall back to the CPU-only index. uv will continue to respect existing index configuration for any packages outside the PyTorch ecosystem.

You can also select a specific backend (e.g., CUDA 12.6) with --torch-backend=cu126 (or UV_TORCH_BACKEND=cu126):

$ # With a command-line argument.
$ uv pip install torch torchvision --torch-backend=cu126

$ # With an environment variable.
$ UV_TORCH_BACKEND=cu126 uv pip install torch torchvision

At present, --torch-backend is only available in the uv pip interface.