Installation
Install OneFlow Stable Version
Install the latest stable version of OneFlow with CUDA support using the following command:
python3 -m pip install -f https://release.oneflow.info oneflow==0.4.0+cu102
Install the latest version of the OneFlow master branch using the following command (not recommended for production environments):
python3 -m pip install oneflow -f https://staging.oneflow.info/branch/master/cu102
If you are informed that the corresponding version cannot be found, please try upgrading pip
:
python3 -m pip install --upgrade pip #--user
Chinese users can use the domestic mirror to accelerate:
python3 -m pip config set global.index-url https://pypi.tuna.tsinghua.edu.cn/simple
Detailed instructions can be found in the pypi mirror help。
System Requirements:
Python >= 3.6
CUDA driver requirements are available in the OneFlow source code repository README
Build from source
If you want to install OneFlow by building from source, please refer to README. Also refer to Troubleshooting for common issues you might encounter when compiling and running OneFlow.
Install OneFlow with legacy CUDA support
To install OneFlow with legacy CUDA support, run one of the following command:
Stable:
python3 -m pip install --find-links https://release.oneflow.info oneflow==0.4.0+[PLATFORM]
Nightly:
python3 -m pip install oneflow -f https://staging.oneflow.info/branch/master/[PLATFORM]
All available [PLATFORM]
:
Platform | CUDA Driver Version | Supported GPUs |
---|---|---|
cu112 | >= 450.80.02 | GTX 10xx, RTX 20xx, A100, RTX 30xx |
cu111 | >= 450.80.02 | GTX 10xx, RTX 20xx, A100, RTX 30xx |
cu110, cu110_xla | >= 450.36.06 | GTX 10xx, RTX 20xx, A100 |
cu102, cu102_xla | >= 440.33 | GTX 10xx, RTX 20xx |
cu101, cu101_xla | >= 418.39 | GTX 10xx, RTX 20xx |
cu100, cu100_xla | >= 410.48 | GTX 10xx, RTX 20xx |
cpu | N/A | N/A |
QQ channel
If you encounter any problems during the installation and want for help, please join the QQ channel or submit issues on Github.
QQ channel ID: 331883 or scan QR code below