We provide pre-built Python wheels for all major OS/PyTorch/CUDA combinations from Python 3.8 till 3.12, see here. Note that currently, Windows wheels are not supported (we are working on fixing this as soon as possible).
To install the wheels, simply run
pip install pyg-lib -f https://data.pyg.org/whl/torch-${TORCH}+${CUDA}.html
where
${TORCH}should be replaced by either1.12.0,1.13.0,2.0.0,2.1.0, or2.2.0${CUDA}should be replaced by eithercpu,cu102,cu113,cu116,cu117,cu118, orcu121
The following combinations are supported:
| PyTorch 2.2 | cpu |
cu102 |
cu113 |
cu116 |
cu117 |
cu118 |
cu121 |
|---|---|---|---|---|---|---|---|
| Linux | ✅ | ✅ | ✅ | ||||
| Windows | |||||||
| macOS | ✅ |
| PyTorch 2.1 | cpu |
cu102 |
cu113 |
cu116 |
cu117 |
cu118 |
cu121 |
|---|---|---|---|---|---|---|---|
| Linux | ✅ | ✅ | ✅ | ||||
| Windows | |||||||
| macOS | ✅ |
| PyTorch 2.0 | cpu |
cu102 |
cu113 |
cu116 |
cu117 |
cu118 |
cu121 |
|---|---|---|---|---|---|---|---|
| Linux | ✅ | ✅ | ✅ | ||||
| Windows | |||||||
| macOS | ✅ |
| PyTorch 1.13 | cpu |
cu102 |
cu113 |
cu116 |
cu117 |
cu118 |
cu121 |
|---|---|---|---|---|---|---|---|
| Linux | ✅ | ✅ | ✅ | ||||
| Windows | |||||||
| macOS | ✅ |
| PyTorch 1.12 | cpu |
cu102 |
cu113 |
cu116 |
cu117 |
cu118 |
cu121 |
|---|---|---|---|---|---|---|---|
| Linux | ✅ | ✅ | ✅ | ✅ | |||
| Windows | |||||||
| macOS | ✅ |
Nightly wheels are provided for Linux from Python 3.8 till 3.12:
pip install pyg-lib -f https://data.pyg.org/whl/nightly/torch-${TORCH}+${CUDA}.html
pip install ninja wheel
pip install git+https://github.com/pyg-team/pyg-lib.git