Yingma’s domestically produced high-end edge computing box is on the market! Equipped with TPU processor BM1684X, compatible with Kirin system and supports OTA upgrade!

As artificial intelligence technology continues to penetrate into practical application scenarios and accelerate the implementation of scenario applications in various industries, the importance of edge computing has become increasingly prominent. Compared with traditional centralized cloud computing, edge computing provides computing capabilities closer to data sources or users, which not only meets the application needs of scenarios with high real-time requirements, but also saves bandwidth, development and operation and maintenance costs. .

Edit switch to center

Add a comment on the image, no more than 140 words (optional)

Over the years, Yingma Technology has been focusing on deeply exploring the potential and demand in the field of edge computing. It is committed to product innovation and technical practice of domestic edge computing, and has launched products with different levels of computing power, multiple platforms, easy deployment, and high reliability for different industries. Domestic edge computing products.

In order to cope with the explosive growth in computing power demand, effectively carry the high-density computing load required for intelligent video analysis, and give edge AI video analysis more excellent performance, Yingma launched the edge based on the fourth generation AI processor BM1684X Computing box - IVP03X intelligent workstation, which has the characteristics of high computing power, strong encoding and decoding capabilities, and adaptability to domestic systems. It can also provide Yingma's self-developed 0-code algorithm transplant tool chain service to meet the needs of intelligent computing and intelligent control in many industries. Application requirements.

Edit switch to center

Add a comment on the image, no more than 140 words (optional)

Yingma Technology IVP03X

Doubled performance, more stable and reliable

Equipped with a new generation of AI processor BM1684X

Yingma IVP03X smart workstation is equipped with a new generation of AI processor BM1684X, 8-core ARM Cortex-A53, clocked at 2.3GHz; INT8 computing power is up to 32Tops, FP16 computing power is up to 16 TFLOPS, FP32 computing power is 2 TFLOPS, which is rare on the edge. Large computing power; supports 32 channels of H.264 & H.265 real-time decoding and 12 channels of H.264 & H.265 encoding, and can process and analyze 16 channels of high-definition videos at the same time. Compared with the previous generation of BM1684 series edge computing boxes, the overall performance of the Yingma IVP03X has been greatly improved, providing a smoother and more efficient application experience for model training and inference in different application scenarios at the edge.

Edit switch to center

Add a comment on the image, no more than 140 words (optional)

Domestic replacement, easier to maintain

Driven by the two-way market and policies, the demand for domestic edge computing products is growing; as the first domestic smart workstation of its kind to support FP16 floating point operations, IVP03X is also the first domestic edge computing box to support OTA upgrade functions, meeting the needs of It can effectively reduce operation and maintenance costs while implementing most edge inference applications.

Edit switch to center

Add a comment on the image, no more than 140 words (optional)

Rich interfaces, ultra-wide temperature design

IVP03X is designed with more than ten IoT application peripheral interfaces such as USB3.0, USB2.0, HDMI, Ethernet, M.2 (SATA), etc. It also supports 4G/5G/WIFI and mSATA expansion, and can be used for edge applications. To provide more convenient link support and meet the needs of larger amounts of data storage; in addition, IVP03X adopts a sub-industrial grade wide temperature design, supporting stable operation in the harsh environment of -40℃~70℃ (typically -20℃~60℃). Its stability and reliability are better.

Edit switch to center

Add a comment on the image, no more than 140 words (optional)

Exquisite design, making full use of old materials

Compared with similar products on the market, all the interfaces of the IVP03X smart workstation adopt a fully embedded design. The sides of the body are flat and have no protruding accessories. It is more friendly to the installation and use in different deployment environments and is more conducive to the operation and maintenance of the equipment. and stable operation. In addition, the IVP03X smart workstation supports the reuse of original equipment and is plug-and-play, making deployment faster and more convenient.

Edit switch to center

Add a comment on the image, no more than 140 words (optional)

Mature and easy-to-use technical support services

Yingma IVP03X smart workstation is adapted to mainstream deep learning frameworks such as Caffe/TensorFlow/MxNet/PyTorch/ONNX/Paddle. It is one of the few edge computing devices in the industry that is compatible with both domestic and foreign deep learning frameworks; it supports Docker container deployment and Python development environment. Provides a deep learning development toolkit, including underlying libraries (ffmpeg, opencv, etc.), inference deployment tools, and a series of software tools, covering model optimization, efficient runtime support and other capabilities required in the neural network inference phase; provides an open source pipeline Develop tool cases that integrate full data process modules such as pull streaming, decoding, inference analysis, and push streaming.

Edit switch to center

Add a comment on the image, no more than 140 words (optional)

In addition, Yingma also provides a one-click zero-code transplantation tool chain, which can realize one-click transplantation of algorithms from GPU to XPU without touching the customer model files. Algorithm adaptation can be achieved in as fast as 3 days. It can greatly shorten the development cycle of enterprises and save communication costs. It also reduces customers' trial and error costs, enabling AI applications to be implemented quickly, easily and cost-effectively.

6 outstanding advantages, leading similar products

Edit switch to center

Add a comment on the image, no more than 140 words (optional)

Yingma IVP03X intelligent workstation can be widely used in AI intelligent video analysis and lightweight model training in the fields of smart cities, smart security, smart communities and smart transportation. In addition to the hard power of the product, Yingma has stable supply capabilities , providing important guarantee for the rapid implementation of customer projects.

Guess you like

Origin blog.csdn.net/weixin_42780701/article/details/132207824