Intel Neural Compute Stick 2 (NCS2) Deep Neural Network Development Tool NCSM2485.DK

  • RS Stock No. 181-1851
  • Mfr. Part No. NCSM2485.DK
  • Brand Intel
Technical Reference
Legislation and Compliance
RoHS Certificate of Compliance
COO (Country of Origin): CN
Product Details

Intel NCS2 Neural Compute Stick 2

The Neural Compute Stick 2 (NCS2) is a USB stick which offers you access to neural network functionality, without the need for large, expensive hardware. It enables you to incorporate computer vision and artificial intelligence (AI) to your IoT and edge devices. Neural network is a way in which we are able to teach machines to learn like humans.
The Intel NCS2 is based on the Intel Movidius™ Myriad™ VPU which has a dedicated hardware accelerator for DNN interference. The NCS2 is supported by the OpenVINO™ Toolkit.
This is the perfect tool for developers, data scientists, industrial engineers, AI engineers and academics.

How does it work?

The Neural Compute Stick 2 has a USB 2.0 connector which is inserted into your host device, for example a PC, laptop or single board computer (SBC). It is a plug-and-play device so you are ready to start prototyping right away.
The OpenVINO™ Toolkit is there to help you accelerate your deep neural network designs and ideas. It is a high performance, high-efficiency, open-source platform which offers compatibility across different platforms.

Features of the NCS2

Latest NCS model from Intel
Plug-and-play USB format
Perfect for Deep Neural Network applications (DNN)
Features the Intel® Movidus™ Myriad™ X VPU
High performance
Low Power

NCS1 vs. NCS2

The Neural Compute Stick 2 is the latest model from Intel. The main difference between the two models is that the NCS1 uses the Myriad 2 VPU while the NCS2 uses the Myriad X VPU. The previous NCS1 version can be found here - (139-3655)

What is it capable of?

The NCS2 development tool is a means to enabling your imagination. The functionality of this tool can be used to recognise faces, objects or even emotions from an image input. This categorisation of data can then be converted into an action.
For example, Australia’s Little Ripper Lifesaver UAV used the Intel NCS to monitor the New South Wales coastline and identify sharks. This data provided immediate danger detection and enabled more response time.

What is the Neural Network?

Neural networks, also referred to as deep learning, is a type of machine learning that enables us to organise and categorise data from image and video analytics. Deep neural networks (DNN) are slightly more complex, with a minimum of 2 layers.
It works by using a series of algorithms which reflect aspects of how our brain functions. They identify correlations in data and use this to group items together. Neural networks are able to organise data that is not labelled or structured beforehand.
An example of neural network is to detect people’s faces from images and video. Going one step further, the neural network could also be used to detect their emotion.

Attribute Value
Kit Name Neural Compute Stick 2 (NCS2)
Kit Classification Development Kit
86 In stock - FREE next working day delivery available
Price Each
£ 68.43
(exc. VAT)
£ 82.12
(inc. VAT)
Per unit
1 +
Related Products
The STEVAL-IDS001Vx board is an evaluation/demonstration tool for ...
The STEVAL-IDS001Vx board is an evaluation/demonstration tool for the SPIRIT1 low-power sub-1GHz ISM band RF devices. SPIRIT1 is fully compatible with the wireless M-Bus protocol. RF Operating Frequency: 315MHz (STEVAL-IDS001V2), 433MHz (STEVAL-IDS001V3), 868MHz (STEVAL-IDS001V4), 915MHz (STEVAL-IDS001V5)STM32L151CBU6 32-bit Cortex-M3 core microcontroller Chip antennaU.FL socket for external antenna USB connector for PC ...
Xilinx Artix-7 FPGA XC7A100T-1CSG324C or XC7A50T-1CSG324C15,850 logic slices, ...
Xilinx Artix-7 FPGA XC7A100T-1CSG324C or XC7A50T-1CSG324C15,850 logic slices, each with four 6-input LUTs and 8 flip-flops (8,150 logic slices for the A7-50T)4,860 Kbits of fast block RAM (2,700 Kbits for the A7-50T)Six clock management tiles, each with phase-locked loop (PLL) (5 clock management tiles for the A7-50T)240 DSP slices (120 ...
ATmega4809 microcontroller One mechanical user switch One yellow ...
ATmega4809 microcontroller One mechanical user switch One yellow user LED32.768k Hz crystal for accurate timing and clock application, ∼20 ppm On-board debugger Board identification in Atmel Studio/Microchip MPLAB XOne green power and status LEDProgramming and debugging Virtual COM port (CDC)Two logic analyzer channels (DGI GPIO)USB powered Adjustable target voltage MIC5353 ...
The eZ430-F2013 is a complete MSP430 development tool ...
The eZ430-F2013 is a complete MSP430 development tool including all the hardware and software to evaluate the MSP430F2013 and develop a complete project in a convenient USB stick form factor. The eZ430-F2013 supports the Code Composer Essentials and IAR Embedded Workbench Integrated Development Environments to provide full emulation with the ...