Vivado ML
Download Vivado ML Standard Edition free. Purchase licensing options for Enterprise Edition start at $2995.
Features
What’s New in 2021.1:
- Improved performance with ML-based algorithms (10% average QoR improvement)​
- Block Design Containers now available ​
- Introducing Vivado® store ​
- Intelligent Design Runs for last-mile timing closure ​
- DFX for Versal now availableВ
Device Support in Production
- Versalв„ўВ AI Core Series: — XCVC1902 and XCVC1802​
- Versal Prime Series: — XCVM1802​
- Virtex UltraScale+ HBM device: ​ XCVU57P
Features
- Vivado ML Standard: The Vivado ML Standard Edition is the FREE version of the revolutionary design suite. It delivers instant access to some basic Vivado features and functionality at no cost.
- Vivado ML Enterprise:В Vivado ML Enterprise Edition is a paid version of the design suite and includes the device support for all Xilinx devices. You can purchase by selecting «Enterprise» from the «Edition» drop-down menu.
Vivado ML Edition Features | Vivado ML Standard Edition | Vivado ML Enterprise Edition | Vivado Lab Edition |
---|---|---|---|
Licensing Option | Free | 30-day Evaluation — Free On Demand on AWS Marketplace NL: $2995 FL: $3595 | В |
Device Support | Limited Xilinx Devices | All Xilinx DevicesВ | В |
Vivado IP Integrator | В | ||
Dynamic Function eXchange | В | ||
Vitis High-Level Synthesis | В | ||
Vivado Simulator | В | ||
Vivado Device Programmer | |||
Vivado Logic Analyzer | |||
Vivado Serial I/O Analyzer | |||
Debug IP (ILA/VIO/IBERT) | В | ||
Synthesis and Place and Route | В | ||
Vitis Model Composer | Buy В NL — $500 FL — $700 | Buy В NL — $500 FL — $700 | В |
Minimum System Memory Recommendations for the Vivado ML Editions
The following tables provide the typical and peak Vivado memory usage per target device. Xilinx recommends to haveВ have at minimumВ enough physical system memory to handle the peak memory usage.
- Memory usage increasesВ with higher LUT and CLB utilization. The numbers below were generated over an average LUT utilization of approximately 75%.В
- The size and complexity of timing constraints directly impact the memory requirements.
- The numbers below were generated using Vivado in scripted batch mode on a single synthesis and implementation run.
- 32-bit machines are not suitable for these devices. В
- Configuration of a Windows 32-bit machine to utilize 3 GB of memory can be found inВ Answer Record 14932.
В | Windows / Linux (64-bit) | |
Device | Typical | Peak |
All devices* | 20 | 32 |
В | Windows / Linux (64-bit) | |
Device | Typical | Peak |
XCKU3P | 7 | 13 |
XCKU5P | 7 | 13 |
XCKU9P | 8 | 13 |
XCKU11P | 9 | 13 |
XCKU13P | 10 | 14 |
XCKU15P | 10 | 15 |
В | Windows / Linux (64-bit) | |
Device | Typical | Peak |
XCVU3P | 11 | 19 |
XCVU5P | 12 | 19 |
XCVU7P | 15 | 24 |
XCVU9P | 20 | 32 |
XCVU11P | 22 | 32 |
XCVU13P | 28 | 47 |
В | Windows / Linux (64-bit) | |
Device | Typical | Peak |
XCZU2EG | 3 | 5 |
XCZU3EG | 4 | 6 |
XCZU4EV | 5 | 8 |
XCZU5EV | 6 | 9 |
XCZU6EG | 7 | 10 |
XCZU7EV | 8 | 11 |
XCZU9EG | 10 | 14 |
XCZU11EG | 11 | 18 |
XCZU15EG | 11 | 18 |
XCZU17EG | 12 | 18 |
XCZU19EG | 14 | 21 |
В | Windows / Linux (64-bit) | |
Device | Typical | Peak |
XCZU21DR | 10 | 14 |
XCZU25DR | 11 | 14 |
XCZU27DR | 13 | 17 |
XCZU28DR | 14 | 17 |
XCZU29DR | 14 | 17 |
В | Windows / Linux (64-bit) | |
Device | Typical | Peak |
XCKU025 | 5 | 7 |
XCKU035 | 5 | 7 |
XCKU040 | 5 | 7 |
XCKU060 | 7 | 11 |
XCKU085В | 9 | 14 |
XCKU095 | 9 | 14 |
XCKU115 | 9 | 14 |
В | Windows / Linux (64-bit) | |
Device | Typical | Peak |
XCVU065 | 7 | 11 |
XCVU080 | 8 | 12 |
XCVU095 | 9 | 14 |
XCVU125 | 10 | 16 |
XCVU160 | 14 | 20 |
XCVU190 | 18 | 24 |
XCVU440 | 32 | 48 |
В | Windows / Linux (64-bit) | |
Device | Typical | Peak |
XC7V585T | 4 | 6 |
XC7V2000T | 10 | 16 |
XC7VX330T | 3 | 5 |
XC7VX415T | 3 | 5 |
XC7VX485T | 4 | 5 |
XC7VX550T | 4 | 6 |
XC7VX690T | 5 | 7 |
XC7VX980T | 7 | 9 |
XC7VX1140T | 8 | 10 |
XC7VH580T | 4 | 6 |
XC7VH870T | 6 | 8 |
В | Windows / Linux (64-bit) | |
Device | Typical | Peak |
XC7K70T | 1.6 | 2.5 |
XC7K160T | 2 | 3 |
XC7K325T | 3 | 4 |
XC7K355T | 3 | 5 |
XC7K410T | 3 | 5 |
XC7K420T | 3 | 5 |
XC7K480T | 4 | 6.5 |
В | Windows / Linux (64-bit) | |
Device | Typical | Peak |
XC7A15T | 2 | 3 |
XC7A35T | 2 | 3 |
XC7A50T | 2 | 3 |
XC7A75T | 2 | 3 |
XC7A100T | 2 | 3 |
XC7A200T | 2.5 | 3.5 |
В | Windows / Linux (64-bit) | |
Device | Typical | Peak |
XC7Z010 | 1 | 1.6 |
XC7Z015 | 1.3 | 1.9 |
XC7Z020 | 1.3 | 1.9 |
XC7Z030 | 1.8 | 2.7 |
XC7Z035 | 3 | 5 |
XC7Z045 | 3 | 5 |
XilinxВ® supports the following operating systems on x86 and x86-64 processor architectures.
- Windows 7 SP1 Professional (64-bit), English/Japanese.В
- Windows 10.0 Fall Creators update (version 1709), 10.0 Version 1803 (64-bit), English/Japanese.
- Red Hat Enterprise Workstation/Server 7.2, 7.3, and 7.4 (64-bit)
- Red Hat Enterprise Workstation 6.6, 6.7, 6.8, and 6.9 (64-bit)
- SUSE Linux Enterprise 11.4 and 12.3 (64-bit)
- CentOS 7.2, 7.3, and 7.4 (64-bit)
- CentOS 6.7, 6.8, and 6.9 (64-bit)
- Ubuntu Linux 16.04.3 LTS (64-bit)В
Note: Please refer to PetaLinux Tools Documentation: Reference Guide (UG1144)В for more information on Installation Requirements for supported Operating Systems with PetaLinux.В
The following table lists architecture support for commercial products in Vivadoв„ў ML Standard versus Vivado ML Enterprise edition. For non-commercial support all Xilinx automotive devices are supported in Vivado ML Standard Edition when available as production devices in the tools.
Device | Vivado ML Standard Edition | Vivado ML Enterprise Edition |
---|---|---|
ZynqВ® | Zynq-7000 SoC Device: |
• XC7Z010, XC7Z015, XC7Z020, XC7Z030, XC7Z007S, XC7Z012S, and XC7Z014S
 • All
• XCZU2EG, XCZU2CG, XCZU3EG, XCZU3CG XCZU4EG, XCZU4CG, XCZU4EV, XCZU5EG, XCZU5CG, XCZU5EV, XCZU7EV, XCZU7EG, and XCZU7CG
• All
• None
• All
• All
• All
• All
• All
• VC1902
• VC1802 Prime Series
• VM1802
Virtex-7 FPGA:
• None
Virtex UltraScale FPGA:
• None
Virtex-7 FPGA:
• All
Virtex UltraScale FPGA:
• All
Virtex UltraScale+ FPGA:
• All
Virtex UltraScale+ HBM:
• All
Virtex UltraScale+ 58G:
• All
KintexВ®-7 FPGA:
• XC7K70T, XC7K160T
Kintex UltraScale FPGA:
• XCKU025, XCKU035
Kintex UltraScale+ FPGA:
• XCKU3P, XCKU5P
Источник