Skip to main content
Link
Menu
Expand
(external link)
Document
Search
Copy
Copied
ONNX Runtime
Install ONNX Runtime
Get Started
Python
C++
C
C#
Java
JavaScript
Objective-C
Julia and Ruby APIs
Windows
Mobile
Web
ORT Training with PyTorch
Tutorials
API Basics
Accelerate PyTorch
PyTorch Inference
Inference on multiple targets
Accelerate PyTorch Training
Accelerate TensorFlow
Accelerate Hugging Face
Deploy on AzureML
Deploy on mobile
Mobile objection detection on iOS
Mobile image recognition on Android
Improve image resolution on mobile
ORT Mobile Model Export Helpers
Deploy on web
Classify images with ONNX Runtime and Next.js
Custom Excel Functions for BERT Tasks in JavaScript
Build a web app with ONNX Runtime
Deploy on IoT and edge
IoT Deployment on Raspberry Pi
Deploy traditional ML
Inference with C#
Inference BERT NLP with C#
Configure CUDA for GPU with C#
Image recognition with ResNet50v2 in C#
Stable Diffusion with C#
Object detection in C# using OpenVINO
Object detection with Faster RCNN in C#
API Docs
Build ONNX Runtime
Build for inferencing
Build for training
Build with different EPs
Build for web
Build for Android
Build for iOS
Custom build
Execution Providers
NVIDIA - CUDA
NVIDIA - TensorRT
Intel - OpenVINO™
Intel - oneDNN
Windows - DirectML
Qualcomm - SNPE
Android - NNAPI
Apple - CoreML
XNNPACK
AMD - MIGraphX
AMD - ROCm
Cloud - Azure
Community-maintained
Arm - ACL
Arm - Arm NN
Apache - TVM
Rockchip - RKNPU
Xilinx - Vitis AI
Huawei - CANN
Add a new provider
Performance
Tune performance
Profiling tools
Memory consumption
Thread management
I/O Binding
Troubleshooting
Model optimizations
Quantize ONNX models
Float16 and mixed precision models
Graph optimizations
ORT model format
ORT model format runtime optimization
Transformers optimizer
Ecosystem
Reference
Releases
Compatibility
Operators
Operator kernels
ORT Mobile operators
Contrib operators
Custom operators
Reduced operator config file
Architecture
Citing ONNX Runtime
ONNX Runtime
Install
Get Started
Tutorials
API Docs
YouTube
GitHub
ONNX Runtime Performance
Table of contents
Tune performance
Model optimizations
Transformers optimizer
Tune Mobile Performance (ORT <1.10 only)