ONNX Runtime is an open-source scoring engine for Open Neural Network Exchange (ONNX) models.
ONNX is an open format for machine learning (ML) models that is supported by various ML and DNN frameworks and tools. This format makes it easier to interoperate between frameworks and to maximize the reach of your hardware optimization investments. Learn more about ONNX on https://onnx.ai or view the Github Repo.
ONNX Runtime is an open architecture that is continually evolving to adapt to and address the newest developments and challenges in AI and Deep Learning. We will keep ONNX Runtime up to date with the ONNX standard, supporting all ONNX releases with future compatibliity while maintaining backwards compatibility with prior releases.
ONNX Runtime continuously strives to provide top performance for a broad and growing number of usage scenarios in Machine Learning. Our investments focus on these 3 core areas:
- Run any ONNX model
- High performance
- Cross platform
ONNX Runtime provides comprehensive support of the ONNX spec and can be used to run all models based on ONNX v1.2.1 and higher. See ONNX version release details here.
As of November 2018, ONNX Runtime supports the latest released version of ONNX (1.3). Once 1.4 is released, ONNX Runtime will align with the updated spec, adding support for new operators and other capabilities.
ONNX Runtime fully supports the ONNX-ML profile of the ONNX spec for traditional ML scenarios.
You can use ONNX Runtime with both CPU and GPU hardware. You can also plug in additional execution providers to ONNX Runtime. With many graph optimizations and various accelerators, ONNX Runtime can often provide lower latency and higher efficiency compared to other runtimes. This provides smoother end-to-end customer experiences and lower costs from improved machine utilization.
Currently ONNX Runtime supports CUDA and MKL-DNN (with option to build with MKL) for computation acceleration. To add an execution provider, please refer to this page.
We are continuously working to integrate new execution providers to provide improvements in latency and efficiency. We have ongoing collaborations to integrate the following with ONNX Runtime: * Intel MKL-DNN and nGraph * NVIDIA TensorRT
ONNX Runtime offers:
- APIs for Python, C#, and C (experimental)
- Available for Linux, Windows, and Mac
See API documentation and package installation instructions below.
Looking ahead: To broaden the reach of the runtime, we will continue investments to make ONNX Runtime available and compatible with more platforms. These include but are not limited to:
- C# for Linux, Mac
- C# supporting GPU
- C packages
- ARM
If you need a model:
- Check out the ONNX Model Zoo for ready-to-use pre-trained models.
- To get an ONNX model by exporting from various frameworks, see ONNX Tutorials.
If you already have an ONNX model, just install the runtime for your machine to try it out. One easy way to deploy the model on the cloud is by using Azure Machine Learning. See detailed instructions here.
| API Documentation | CPU package | GPU package |
|---|---|---|
| Python | Windows Linux Mac |
Windows Linux |
| C# | Windows Linux - Coming Soon Mac - Coming Soon |
Coming Soon |
| C (experimental) | Coming Soon | Coming Soon |
| C++ | TBD | TBD |
For details on the build configurations and information on how to create a build, see Build ONNX Runtime.
See more details on API and ABI Versioning and ONNX Compatibility in Versioning.
For an overview of the high level architecture and key decisions in the technical design of ONNX Runtime, see Engineering Design.
ONNX Runtime is built with an extensible design that makes it versatile to support a wide array of models with high performance.
- Add a custom operator/kernel
- Add an execution provider
- Add a new graph transform
- Add a new rewrite rule
We welcome your contributions! Please see the contribution guidelines.
For any feedback or to report a bug, please file a GitHub Issue.
This project has adopted the Microsoft Open Source Code of Conduct. For more information see the Code of Conduct FAQ or contact [email protected] with any additional questions or comments.