DeepSpeed vs ONNX Runtime
DeepSpeed
deepspeed.aiONNX Runtime
onnxruntime.aiSide by side
| Figure | DeepSpeed | ONNX Runtime |
|---|---|---|
| Price from | Free | Free |
| Free plan | Yes | Yes |
| Free trial | — | — |
| Open source | — | — |
| Web | — | Yes |
| Windows | — | Yes |
| macOS | Yes | Yes |
| Linux | Yes | Yes |
| iOS | — | Yes |
| Android | — | Yes |
| Self-hosted | Yes | Yes |
| Plans published | 1 | 1 |
| Spec score · Deep Learning Software | 8.0 (#12) | 8.4 (#6) |
| Integrations | The site lists integrations with Hugging Face Transformers, Accelerate, PyTorch Lightning and MosaicML. | The ecosystem documentation lists integrations with Azure Machine Learning, Azure Custom Vision, Azure SQL Edge, Azure Synapse Analytics, ML.NET, and NVIDIA Triton Inference Server. |
| Purpose | DeepSpeed is a deep learning optimization library for distributed model training and inference. | ONNX Runtime is a cross-platform machine-learning model accelerator with interfaces for hardware-specific libraries. |
| Support | The GitHub repository says DeepSpeed holds public office hours on the last Tuesday of each month. | Documentation questions are directed to issue filing, and the project invites users to report bugs, suggest features, and submit pull requests on GitHub. |
| Training | Its training features include mixed precision, data, model and pipeline parallelism, and the ZeRO optimizer. | ONNX Runtime supports large-model training and on-device training for personalization and federated-learning scenarios. |
Plans
In their own words
ONNX Runtime
ONNX Runtime is a free engine for machine-learning inference and training within existing technology stacks. It can run models from frameworks including PyTorch, TensorFlow/Keras, TFLite, and scikit-learn. For inference, it optimizes latency, throughput, memory use, and binary size through graph transformations, accelerator-aware partitioning, and optimized computation kernels. Its Execution Providers connect ONNX models to hardware acceleration libraries for CPUs, GPUs, FPGAs, and specialized NPUs. Listed options include NVIDIA CUDA and TensorRT, Intel OpenVINO, DirectML, Qualcomm QNN, Android NNAPI, Apple CoreML, and WebGPU. Deployment spans cloud servers, edge and mobile devices, and browsers; its web runtime supports browser models and its mobile runtime supports Android and iOS. Supported languages include Python, C#, C++, Java, JavaScript, and Rust. Developers can create smaller web or mobile packages by including only the operators and opsets their models need. The open-source runtime is available on Android, iOS, Linux, macOS, self-hosted environments, web, and Windows.