May 22, 2023
press release
News highlights:
- The first silicon of Supergate’s deep learning microcontroller (DMC) was received in February 2023, and the test network was immediately up and running successfully on the NMP-300 IP licensed from Aim Future.
- An on-floor demonstration showcases ultra-low-power object detection for consumer electronics.
- DMC is the perfect solution for a wide range of applications such as home consumer, robotics, smart toys, and artificial intelligence of things (AIoT).
- Products will be available for purchase from August 2023.
Seoul, South Korea and San Jose, California. May 22, 2023 – SUPERGATE today announced that its latest product, a Deep Learning Microcontroller (DMC), will be demonstrated at the 2023 Embedded Vision Summit, the premier event for practical and deployable computer vision and visual AI. Did. Bring visual intelligence to your products. This year’s summit will be held May 22-24, 2023 in Santa Clara, California. This annual event brings together technology professionals from around the world from companies developing computer vision and edge AI-enabled products, including embedded systems, cloud solutions and mobile applications. Supergate also announced that it has licensed the NMP-300 processor from AiM Future. The NMP-300 is the smallest member of the NeuralMosAIc processor family, enabling ultra-low-power microcontrollers for intelligent sensor and control systems in a variety of applications such as home consumer, robotics, toys, and pervasive artificial intelligence (AIoT) optimized for ). Supergate will be showcasing a DMC-based computer vision demo at booth #801, and AiM Future will be showcasing concurrent multimodal inference of objects on NeuroMosAIc processors at booth #802.
“Our customers continue to redefine consumer expectations by providing sophisticated interactive devices that stretch the imagination,” said Seungpil Shim, Chief Executive Officer of Supergate. “AiM Future has proven to be a strategic partner enabling DMC to address intelligent end-device applications enabled by NeuroMosAIc processor technology. Underneath, we’re introducing a new level of on-device AI capabilities never seen before.”
“We are delighted that Supergate has chosen to partner with AiM Future on this innovative new deep learning microcontroller product. We have licensed the NMP-300 as a proven solution in ,” said ChangSoo Kim, CEO of AiM Future. “The NMP-300’s small size and low power consumption make it an ideal fit for the applications targeted by his DMC at Supergate, resulting in compelling advantages over traditional microcontrollers.”
DMC is a 32-bit RISC-V based microcontroller for low-cost AI inference applications. DMC integrates the highly optimized NMP-300 for neural networks in general and convolutional neural networks (CNNs) in particular. DMC is ideal for Tiny Machine Learning (TinyML) and Artificial Intelligence of Things (AIoT) applications. Combining TinyML and AIoT enables the development of intelligent connected devices that can perform complex tasks with minimal power consumption and latency. For example, TinyML models can be deployed in low-power microcontrollers in smart home devices such as thermostats to optimize temperature control based on user preferences and patterns. The device can be connected to an AIoT platform that collects data from multiple devices in the home and uses machine learning algorithms to optimize energy usage and improve overall efficiency.
The NMP-300 is ideal for machine learning applications in ultra-low power far-edge devices such as wearables, environmental sensors, and image sensors. It delivers 0.5 TOPS while consuming microwatts in common sensor applications.
The NMP-500 targets applications that require performance efficiency, such as smartphones, drones, AR/VR, and high-end consumer electronics. It offers up to 4 TOPS, but can be scaled down to 2 TOPS to meet power and area constraints in cost-sensitive applications.
The NMP-700 is the highest performing member of the family for edge gateways, servers, robotics and UAVs. Achieve 8 or 16 TOPS when performance is the primary design goal.
The co-designed NeuroMosAIc Studio provides developers with a comprehensive set of tools for running pre-trained machine learning models on NMP hardware. It includes advanced hardware-aware model converters, mappers and profilers for maximum efficiency. Support for industry-standard frameworks such as ONNX, PyTorch, and TensorFlow Lite provides flexibility for ML engineers, while leveraging the GLOW and TensorFlow XLA compilers for highly optimized models. It represents a production-ready hardware and software solution that produces the highest performance and lowest power calculations for edge AI applications.
- DMC is now available for sampling by major customers.
- Please contact us for more information. [email protected].
- NeuroMosAIc Processor IP and Studio are now commercially available.
- Contact us for more information [email protected].
Supergate is a turnkey design specialist that offers wafers or chips with unique SoC frameworks and solutions covering autonomous vehicles, high performance computing and artificial intelligence of things. For more information, please visit http://www.supergate.cc.
Based in Seoul, South Korea, the company was founded at the end of 2020 with LG Electronics as a strategic investor. It has since secured a seed round investment from a group of leading venture capital firms We Ventures, KB Investment and D.Camp. The company focuses on developing neural network hardware accelerator IP and compiler software. Its flagship product, his NeuroMosAIc architecture, was commercialized in his 2019, and the company has multiple licensing agreements with partners around the world. For more information, please visit https://www.aimfuture.ai.
