
The Xilinx VCK5000 Versal development card is built on the Xilinx 7nm Versal® ACAP architecture and is designed for (AI) Engine development with Vitis end-to-end flow and AI Inference development with partner solutions. * Limited to AI Engine only. For full Vivado® flow and device customization, please contact sales.
The Xilinx® VCK5000 Versal® development card is built on the Xilinx 7nm Versal ACAP architecture and is designed for 5G, DC compute, AI, Signal Processing, Radar, and other applications. The Domain-Specific Architecture brings strong horsepower per watt while keeping ease-of-use in mind with C/C++, software programmability.
Delivering the world’s first zero dark silicon (near 100% compute efficiency per watt) in standard AI benchmarks and 2x TCO compared to the flagship nVidia GPUs, the VCK5000 is an ideal development platform for CNN, RNN, and NLP acceleration for your cloud and edge applications.
VCK5000 is fully supported by the Vitis™ unified software platform, the Vitis AI™ development environment, and partner AI solutions such as Mipsology and Aupera. If you are an AI developer, bring your TensorFlow / Pytorch trained models to directly infer on Versal using Vitis AI or Mipsology Zebra. If you are looking to implement algorithm acceleration with AI engine and programmable logic, we provide AI engine C/C++ high-level abstraction APIs and Vitis Accelerated Libraries. The Vitis end-to-end flow is developed using C/C++ to run on X86 or embedded processors and manage runtime interactions with the accelerator with XRT. The hardware component, or kernel, can be developed using C/C++, or RTL target on PL or AI Engines.
Explore partner solutions and articles, and learn about the key features for AI Inference Development with the VCK5000
ML Heavy: H.264 Decode + Yolov3 + 3x ResNet-18
Video Heavy: H.264 Decode + tinyYolov3 + 3x ResNet-50
Zebra is the ideal compute engine to accelerate CNN inference. Zebra seamlessly replaces CPUs/GPUs to compute any neural network on an FPGA faster, with lower power consumption, and at a lower cost. Zebra is ‘plug and play’, no knowledge of FPGA technology, compilation, or any changes to the environment or the application is required. Zebra lets AI engineers focus on application development while enjoying unmatched inference performance.
Click on Get Started to see an interactive demo that allows users to see live results of Zebra computing neural network inference based on Xilinx hardware.
This interactive demo allows users to see live results of Zebra computing neural network inference based on Xilinx hardware. Users can select the neural network and the Xilinx hardware and see live results. Each result can be inspected directly in a web browser for each image processed.
Step 1: Click the Zebra performance demo link
Step 2: Select your neural network
Step 3: Select your Xilinx hardware
Step 4: See live results of Zebra inference computation using your selections.
Step 5: Fill out the Zebra inquiry form to get information on how to download Zebra
Step 1: Try Zebra VCK5000 Live demo link
Step 2: Drag and drop your photos in the white text box on the top of the screen
Step 3: See live results of Zebra inference computation on the VCK5000 using your selections.
Step 4: Fill out the Zebra inquiry form to get information on how to download Zebra
Aupera VMSS solutions support a high density of video feeds from multiple full HD cameras for object identification and classification. The additional unique proposition of this solution is its ability to run multiple inferencing models concurrently and deliver accurate results at a deterministic low latency, resulting in industry-leading low Total Cost of Ownership (TCO)
Please fill out this form for further information. Our team of specialists is here to support your needs, anywhere in the world.
Data centers are increasingly turning to artificial intelligence to manage various tasks from monitoring equipment to server optimization. At the heart of the data center, FPGA-based adaptive computing is proving itself to be, in many cases, the most efficient and cost-effective solution for running complex AI workloads.
Here are the best uses of the VCK5000 developer card combined with Vitis AI from our 2021 Adaptive Computing Challenge
Get Started with AI Engine Development. Explore Key Features and Articles
Purchase VCK5000 production silicon-based card
Request access to the VCK5000 Versal Development Card Secure Site
Follow the getting started steps and installation guide in the VCK5000 Versal Development Card Secure Site
Card Specifications | VCK5000 | |
---|---|---|
Device | VC1902 | |
Compute | Active | Passive* |
INT8 TOPs (peak) | 145 | 145 |
Dimensions | ||
Height | Full | Full |
Length | Full | 3/4 |
Width | Dual Slot | Dual Slot |
Memory | ||
Off-chip Memory Capacity | 16 GB | 16 GB |
Off-chip Total Bandwidth | 102.4 GB/s | 102.4 GB/s |
Internal SRAM Capacity | 23.9 MB | 23.9 MB |
Internal SRAM Total Bandwidth | 23.5 TB/s | 23.5 TB/s |
Interfaces | ||
PCI Express | Gen3 x 16 / Gen4 x 8 | Gen3 x 16 / Gen4 x 8 |
Network Interfaces | 2x QSFP28 (100GbE) | 2x QSFP28 (100GbE) |
Logic Resources | ||
Look-up Tables (LUTs) | 899,840 | 899,840 |
Power and Thermal | ||
Maximum Total Power | 225W | 225W |
Thermal Cooling | Active | Passive |
* We will ship the Active board only. If you remove the fans from the VCK5000, following the Hardware Installation Guide, it becomes Passive.