Step 1: Download and install Vitis AI from Github
The Vitis™ AI software is made available via Docker Hub. Vitis AI consists of the following two packages:
Vitis AI Tools Docker | Vitis AI runtime package for edge
Step 2: Hardware platform setup
Embedded SoC: ZCU102/ZCU104 set up.
Alveo: U200/U250 set up | U50/U280 set up
Step 3: Run Vitis AI examples
Embedded SoC: ZCU102/ZCU104
Alveo: U200/U250 | U50/U280
Nimbix Cloud
Evaluate Vitis AI with Alveo accelerator cards on Nimbix. Quickly evaluate the performance benefits Xilinx platforms can bring to your applications, with no upfront purchase of cards or local software setup required.
Learn More >
In Depth Tutorials
Vitis AI Introduction Tutorial for the ZCU104
This tutorial puts in practice the concepts of FPGA acceleration of Machine Learning and illustrates how to quickly get started deploying both pre-optimized and customized ML models on Xilinx devices
Vitis AI Custom Embedded Platform Creation
In this module, we will create a custom Vitis embedded platform for ZCU104. It will be capable to run Vitis acceleration applications including Vitis-AI applications. Of course, general embedded software application can also run on this platform.
Intelligent Video Analytics ZCU104 ML Acceleration
This repository contains a reference design for the use of the Xilinx IVAS framework in a multi-channel machine learning system.
Get technical resources for developers as well as code samples for Vitis AI
Vitis AI Product Documentation
Curated articles form industry experts for all things Vitis AI
View all Vitis AI Articles >
Access free Vitis AI training course when you sign up for the Developer Program Secure site.
Courses include:
Introduction to the Vitis AI Development Environment
Frameworks Supported by the Vitis AI Development Environment
Setting Up the Vitis AI Development Environment
You can also visit our Xilinx Customer Training Center for additional Paid Courses
Watch various videos such as quick-take product introductions, tutorial walk-throughs and demos.
Introduction to Vitis AI
Vitis AI empowers software developers to keep up with AI innovation and unifies AI application development from edge to cloud. Learn more and download today.
Real-time Multi-class 3D Object Detection
With end-to-end HW/SW co-optimization through Vitis AI, we run real-time multiclass 3D detection using the state-of-the-art PointPillars model on Xilinx ZU+ MPSoC.
IBM PowerAI Vision Inference Acclerated by Xilinx Vitis AI
Accelerate your inference using the Xilinx Alveo U50 Adaptable Accelerator Card with IBM PowerAI Vision
Edge-to-Cloud Video Analytics
Demonstration of how Xilinx can accelerate both the development and deployment of real-time HD video applications with machine learning. Showcases video processing and machine learning in a single system using industry-standard frameworks FFmpeg and YOLO.
DPU-PYNQ for Python-Powered Edge AI Appliances
Chris Anderson chats with Quenton Hall of Xilinx about how developers can leverage ZYNQ FPGAs in Edge AI appliances.
Adaptable AI Inference with Vitis AI
In this webinar, go in-depth with the key components of Vitis AI and learn how to achieve adaptable and efficient AI inference on Xilinx hardware platforms.
Vitis AI Deep Dive
In this Webinar, learn to use Vitis AI to deploy and run your pre-trained DNN models to Xilinx’s embedded SoC and Alveo acceleration platforms. Then get started with using Vitis AI to run examples on the board.
Accelerating AI Camera Development with Xilinx Vitis
Learn how to leverage Xilinx MPSoCs with Vitis in order to implement AI Camera designs
Learn how developers are using Xilinx technologies to accelerate their work. Learn from the tutorials, articles, and projects from the community.
Share your work (Github repo, Hackster.io link etc) with us at developer@xilinx.com and every month we will select projects to be featured on our developer site. Projects with be based on design features, performance, creativity, and originality.