Adaptive
Inference Acceleration
4x Faster than GPUs
Fastest Real-Time Inference
For low-latency ML inference, Xilinx delivers leadership throughput and power efficiency. In standard benchmark tests on GoogleNet V1, Xilinx U250 delivers more than 4x the throughput of the fastest GPU at real-time inference. Learn more in the whitepaper: “Accelerating DNNs with Xilinx Alveo Accelerator Cards”.
xDNNv3 will be available in November 2018, in the ML Suite. Get started today with the ML Suite featuring xDNNv2 from the links below.
Learn More
* See White Paper for performance details
Whole Application Acceleration
Xilinx ML Suite
The Xilinx ML Suite enables developers to optimize and deploy accelerated ML inference. It provides support for many common machine learning frameworks such as Caffe, MxNet and Tensorflow as well as Python and RESTful APIs. Suite components:
- xfDNN compiler/optimizer – auto-layer fusing, memory optimization, and framework integration
- xfDNN quantizer – improves performance with auto model-precision INT8 calibration
- Platforms – deployable on-premise or through cloud services
- Deployable on-premise or through cloud services: Amazon, Nimbix, Huawei, Alibaba Cloud, Baidu, Tencent
Xilinx ML Acceleration
on AWS
Get Started
Xilinx ML Acceleration
on Nimbix
Get Started
Data Center Acceleration
on Alveo
Get Started
Xilinx DeePhi Edge Platform
ML Inference performance leadership with CNN pruning technology.
- 5X to 50X network performance optimization
- Increases FPS and reduces power
Optimization/Acceleration Complier Tools
- Supports networks from Tensorflow, Caffe,and MXNet
- Compiles networks to optimized Xilinx MPSoC accelerator runtime
Get more information
about DeePhi
Learn More
Watch a Demo about
DeePhi Edge Platform
Watch Video
Join the DeePhi
Community Forum
Join Now
Documentation
Resources
Industry and Academia
Acceleration Knowledge Center
Documentation, Resources, Papers, Tutorials for Cloud Inference
Learn MoreXilinx University Program
Enabling the use of Xilinx technologies for academic teaching and research
Learn More