Hardware: Modules, Cards & Systems—Purchase

SAKURA-II M.2 8GB Module

The small form factor M.2 Module, featuring the SAKURA-II accelerator, is ideal for space-constrained designs and features significant onboard DRAM, making it optimal for Generative AI and other memory-intensive AI applications like Large Language Models (LLMs).

	DRAM Memory: 8GB (2x banks of 4GB LPDDR4)
	Form Factor: M.2 Key M 2280 (22mm x 80mm)
	Host Interface: PCIe Gen 3.0 x4
	Peak Performance: 60 TOPS with INT8, 30 TFLOPS with BF16
	Module Power: 10W (typical)

Best choice for space-constrained and power limited designs with easy integration into small edge AI systems!

SAKURA-II M.2 Eval Module (16GB)

	DRAM Memory: 16GB (2x banks of 8GB LPDDR4)
	Form Factor: M.2 Key M 2280 (22mm x 80mm)
	Host Interface: PCIe Gen 3.0 x4
	Peak Performance: 60 TOPS with INT8, 30 TFLOPS with BF16
	Module Power: 10W (typical)
	Included Hardware: Fan sink

Best choice for space-constrained and power-limited designs with easy integration into small edge AI systems!

SAKURA-II M.2 Module

The compact M.2 Module, equipped with the high-performance SAKURA-II accelerator with up to 60 TOPS, is designed for space-constrained environments and features 16GB onboard DRAM. This makes it particularly suitable for Generative AI and other memory-intensive AI applications such as Large Language Models (LLMs). The module can be inserted into M.2 slots for immediate evaluation.

It is the optimal choice for designs that require both space efficiency and power limitations, facilitating seamless integration into small edge AI systems.

Get a pre-built system with the SAKURA-II M.2 Module, ready for development!

Large DRAM Capacity: 16GB of LPDDR4 DRAM, enabling efficient processing of complex vision and Generative AI workloads	Enhanced Memory Bandwidth: Up to 4x more DRAM bandwidth than competing AI accelerators, ensuring superior performance for LLMs and LVMs
16GB (2 banks of 8GB LPDDR4)	Up to 68 GB/sec
High Performance: SAKURA-II edge AI accelerator running the latest AI models	Low Power: Optimized for low power even during high utilization
60 TOPS (INT8), 30 TFLOPS (BF16)	10W (typical)
Form Factor:	Module Dimensions:
M.2 Key-M 2280	M.2 2280-D6: 22.0mm (W), 80.0mm (L), 3.2mm (T), 1.5mm (B) With fan sink: 24.5mm (W), 80.0mm (L), 25.2mm (T), 2.5mm (B)
Host Interface:	Operating Temperature Range (T_j):
PCIe Gen 3.0 x4	-20C to 85C (non-condensing)

EdgeCortix-S2LP-Single-short-bracket-tilt

SAKURA-II Single Eval Card (16GB)

The PCIe Card featuring the SAKURA-II accelerator, is ideal for insertion into PCIe backplanes and fits comfortably into a single slot even with the attached heat sink. This card provides significant onboard DRAM, making it optimal for Generative AI and other memory-intensive AI applications like Large Language Models (LLMs).

	DRAM Memory: 16GB (2x banks of 8GB LPDDR4)
	Form Factor: PCIe Half Height, Half Length (HHHL) slot, x16 mechanical
	Host Interface: PCIe Gen 3.0 x8
	Peak Performance: 60 TOPS with INT8, 30 TFLOPS with BF16
	Card Power: 10W (typical)
	Included Hardware: Half-height, full-height brackets and heat sink

Best choice for most Edge AI designs using the industry standard PCIe form factor!

SAKURA-II Single PCIe Card

SAKURA-II Single PCIe Card is a high-performance edge AI accelerator with up to 60 TOPS in a low-profile single-slot PCIe form factor. It features 16GB onboard DRAM, ideal for Generative AI models and memory-intensive applications like Large Language Models.

The card adheres to PCIe specifications and fits into a single slot with a passive heat sink, occupying a single Half Height, Half Length (HHHL) ;x16 slot.

Available in configurations with one or two SAKURA-II AI Accelerators.

Upgrade to the Dual PCIe Card for maximum AI processing power!

Large DRAM Capacity: Up to 16GB of LPDDR4 DRAM, enabling efficient processing of complex vision and Generative AI workloads	Low Power: Optimized for low power while processing AI workloads with high utilization
16GB - 2 banks, 8GB LPDDR4	10W typical
High Performance: SAKURA-II edge AI accelerator running the latest AI models	Interface:
60 TOPS (INT8), 30 TFLOPS (BF16)	PCIe Gen 3.0 x8 interface (x16 mechanical) USB-C: Utility and debug port
Enhanced Memory Bandwidth: Up to 4x more DRAM bandwidth than competing AI accelerators	Form Factor:
Up to 68 GB/sec	Half Height, Half Length (HHHL) x16 single-width slot card
Included Hardware:	Operating Temperature Range (T_j):
Half and full-height brackets Passive heat sink (assembled)	-20C to 85C (non-condensing)

SAKURA-II Dual Eval Card (32GB)

The Dual PCIe Card, featuring two SAKURA-II accelerators, is ideal for insertion into PCIe backplanes and fits comfortably into a single slot even with the attached heat sink. This card provides significant onboard DRAM, making it optimal for Generative AI and other memory-intensive AI applications like Large Language Models (LLMs).

	DRAM Memory: 32GB (4x banks of 8GB LPDDR4)
	Form Factor: Half Height, Half Length (HHHL) x16 slot card
	Host Interface: PCIe Gen 3.0 x8/x8 (bifurcated)
	Peak Performance: 120 TOPS with INT8, 60 TFLOPS with BF16
	Card Power: 20W (typical)
	Included Hardware: Half-height, full-height brackets and heat sinks

Best choice for designs that need the highest AI processing capability!

SAKURA-II Dual PCIe Card

The SAKURA-II Dual PCIe Card is a high-performance edge AI accelerator, delivering up to 120 TOPS in a single-slot PCIe form factor. With 32GB of onboard DRAM, it’s ideal for large Generative AI models and AI applications with high memory demands, such as Large Language Models.

This card fits into PCIe slots for immediate use and occupies a single Half Height, Half Length (HHHL) x16 slot. The dual PCIe configuration includes a second SAKURA-II, and doubles the DRAM in a single slot, offering flexibility to use the accelerators separately or combined for enhanced AI inferencing.

Get a pre-built system with the SAKURA-II Card, ready for development!

Large DRAM Capacity: Up to 32GB of LPDDR4 DRAM, enabling efficient processing of complex vision and Generative AI workloads	Low Power: Optimized for low power while processing AI workloads with high utilization
32GB - 4 banks 8GB LPDDR4	20W typical
High Performance: SAKURA-II edge AI accelerator running the latest AI models	Interface:
120 TOPS (INT8), 60 TFLOPS (BF16)	PCIe Gen 3.0 bifurcated x8/x8 interface (one for each SAKURA-II device) USB-C: Utility and debug port
Enhanced Memory Bandwidth: Up to 4x more DRAM bandwidth than competing AI accelerators	Form Factor:
Up to 68 GB/sec	Half Height, Half Length (HHHL) x16 single-width slot card
Included Hardware:	Temperature Range:
Half and full-height brackets Passive heat sink (assembled)	-20C to 85C (non-condensing)

SAKURA-II M.2 System Overview

Optimized for Generative AI Model Deployment

The SAKURA-II M.2 Development System is a high-performance setup designed for advanced AI modeling and computation tasks. Powered by a Ryzen 7 7700 processor, it ensures exceptional processing speed and efficiency. The system features the SAKURA-II M.2 Module with 16GB capacity, providing robust performance for intensive workloads.

Additional configuration includes 64GB of DDR5-5600 RAM, delivering high-speed memory support essential for running complex AI algorithms. A 500GB SSD ensures fast storage and retrieval of data, enabling seamless model training and deployment. Running on Ubuntu, this system is meticulously configured for AI model development and execution, making it a reliable choice for cutting-edge projects.

The SAKURA-II M.2 Development System offers up to 60 TOPS (INT8) and 30 TFLOPS (BF16) for efficient AI computation. Its compact M.2 Key M 2280 form factor and PCIe Gen 3.0 x4 interface ensure robust connectivity and suitability for demanding AI workloads.

SAKURA-II Single PCIe System Overview

Optimized for Generative AI Model Deployment

The SAKURA-II Single PCIe Development System is a high-performance setup designed for advanced AI modeling and computation tasks. Powered by a Ryzen 7 7700 processor, it ensures exceptional processing speed and efficiency. The system features the SAKURA-II Single PCIe Card with 16GB capacity, providing robust performance for intensive workloads.

The SAKURA-II Single PCIe Development System offers up 60 TOPS (INT8) and 30 TFLOPS (BF16) for efficient AI computation. Its compact PCIe Half-Height, Half-Length (HHHL) x16 form factor and PCIe Gen 3.0 x8 interface ensure robust connectivity and suitability for demanding AI workloads.

Upgrade to the Dual PCIe pre-built system for maximum AI processing power, ready for development!

SAKURA-II Dual PCIe System Overview

Maximized for Generative AI Model Deployment

The SAKURA-II Dual PCIe Development System is a high-performance setup designed for advanced AI modeling and computation tasks. Powered by a Ryzen 7 7700 processor, it ensures exceptional processing speed and efficiency. The system features the SAKURA-II Dual PCIe Card with 32GB capacity, providing an even more robust performance for intensive workloads.

The SAKURA-II Dual PCIe Development System offers up to 120 TOPS (INT8) and 60 TFLOPS (BF16) for efficient AI computation. Its compact PCIe Half-Height, Half-Length (HHHL) x16 form factor and PCIe Gen 3.0 x8/x8 interface ensure robust connectivity and suitability for highly demanding AI workloads.

SAKURA^®-II

SAKURA-II AI Accelerator Modules, Cards, and Development Systems

SAKURA-II Modules & Cards

SAKURA-II Development Systems

SAKURA-II M.2 8GB Module

$249*

SAKURA-II M.2 Eval Module (16GB)

$349*

SAKURA-II M.2 Module

SAKURA-II Single Eval Card (16GB)

$499*

SAKURA-II Single PCIe Card

SAKURA-II Dual Eval Card (32GB)

$799*

SAKURA-II Dual PCIe Card

SAKURA-II M.2 Development System

$1699*

SAKURA-II M.2 System Overview

Optimized for Generative AI Model Deployment

SAKURA-II Single PCIe Development System

$1799*

SAKURA-II Single PCIe System Overview

Optimized for Generative AI Model Deployment

SAKURA-II Dual PCIe Development System

$1999*

SAKURA-II Dual PCIe System Overview

Maximized for Generative AI Model Deployment

Please fill out for SAKURA-II order inquiry.

Sign up for our latest updates, news and more.

SAKURA®-II

SAKURA-II AI Accelerator Modules, Cards, and Development Systems

SAKURA-II Modules & Cards

SAKURA-II Development Systems

SAKURA-II M.2 8GB Module

$249*

SAKURA-II M.2 Eval Module (16GB)

$349*

SAKURA-II M.2 Module

SAKURA-II Single Eval Card (16GB)

$499*

SAKURA-II Single PCIe Card

SAKURA-II Dual Eval Card (32GB)

$799*

SAKURA-II Dual PCIe Card

SAKURA-II M.2 Development System

$1699*

SAKURA-II M.2 System Overview

Optimized for Generative AI Model Deployment

SAKURA-II Single PCIe Development System

$1799*

SAKURA-II Single PCIe System Overview

Optimized for Generative AI Model Deployment

SAKURA-II Dual PCIe Development System

$1999*

SAKURA-II Dual PCIe System Overview

Maximized for Generative AI Model Deployment

Please fill out for SAKURA-II order inquiry.

SAKURA^®-II