FastDeploy provides pre-built libraries for developers to download and install directly. Meanwhile, FastDeploy also offers easy access to compile so that developers can compile FastDeploy according to their own needs.
This article is divided into two parts:
- CUDA >= 11.2
- cuDNN >= 8.0
- python >= 3.6
- OS: Linux(x64)/Windows 10(x64)
FastDeploy supports Computer Vision, Text and NLP model deployment on CPU and Nvidia GPU with Paddle Inference, ONNX Runtime, OpenVINO and TensorRT inference backends.
Install the released version（the newest 0.7.0 for now）
pip install fastdeploy-gpu-python -f https://www.paddlepaddle.org.cn/whl/fastdeploy.html
Install the Develop version（Nightly build）
pip install fastdeploy-gpu-python==0.0.0 -f https://www.paddlepaddle.org.cn/whl/fastdeploy_nightly_build.htmlWe recommend users to use Conda to configure the development environment.
conda config --add channels conda-forge && conda install cudatoolkit=11.2 cudnn=8.2
Install the released version（Latest 0.7.0）
| Platform | File | Description |
|---|---|---|
| Linux x64 | fastdeploy-linux-x64-gpu-0.7.0.tgz | g++ 8.2, CUDA 11.2, cuDNN 8.2 |
| Windows x64 | fastdeploy-win-x64-gpu-0.7.0.zip | Visual Studio 16 2019, CUDA 11.2, cuDNN 8.2 |
Install the Develop version（Nightly build）
| Platform | File | Description |
|---|---|---|
| Linux x64 | fastdeploy-linux-x64-gpu-0.0.0.tgz | g++ 8.2, CUDA 11.2, cuDNN 8.2 |
| Windows x64 | fastdeploy-win-x64-gpu-0.0.0.zip | Visual Studio 16 2019, CUDA 11.2, cuDNN 8.2 |
- python >= 3.6
- OS: Linux(x64/aarch64)/Windows 10 x64/Mac OSX(x86/aarm64)
FastDeploy supports computer vision, text and NLP model deployment on CPU with Paddle Inference, ONNX Runtime, OpenVINO inference backends. It should be noted that under Linux aarch64 and Mac OSX, only the ONNX Runtime is supported for now.
Install the released version（Latest 0.7.0 for now）
pip install fastdeploy-python -f https://www.paddlepaddle.org.cn/whl/fastdeploy.html
Install the Develop version（Nightly build）
pip install fastdeploy-python==0.0.0 -f https://www.paddlepaddle.org.cn/whl/fastdeploy_nightly_build.htmlInstall the released version（Latest 0.7.0 for now, Android is 0.4.0 pre-release）
| Platform | File | Description |
|---|---|---|
| Linux x64 | fastdeploy-linux-x64-0.7.0.tgz | g++ 8.2 |
| Windows x64 | fastdeploy-win-x64-0.7.0.zip | Visual Studio 16 2019 |
| Mac OSX x64 | fastdeploy-osx-x86_64-0.7.0.tgz | clang++ 10.0.0 |
| Mac OSX arm64 | fastdeploy-osx-arm64-0.7.0.tgz | clang++ 13.0.0 |
| Linux aarch64 | - | - |
| Android armv7&v8 | fastdeploy-android-0.4.0-shared.tgz | NDK 25, clang++, support arm64-v8a及armeabi-v7a |
Install the Develop version（Nightly build）
| Platform | File | Description |
|---|---|---|
| Linux x64 | fastdeploy-linux-x64-0.0.0.tgz | g++ 8.2 |
| Windows x64 | fastdeploy-win-x64-0.0.0.zip | Visual Studio 16 2019 |
| Mac OSX x64 | fastdeploy-osx-arm64-0.0.0.tgz | - |
| Mac OSX arm64 | fastdeploy-osx-arm64-0.0.0.tgz | clang++ 13.0.0编译产出 |
| Linux aarch64 | - | - |
| Android armv7&v8 | - | - |