SAN JOSE, Calif.--(BUSINESS WIRE)--GrAI Matter Labs (GML) today unveiled a hardware development kit (HDK) for GrAI One, the world’s first AI processor optimized for ultra-low latency and low power processing at the edge.
“Today over 95% of compute cycles consumed at the edge for real-time AI applications are wasted, since current solutions cannot leverage time-based sparsity. GrAI One changes that.”
GML is making the GrAI One HDK available for companies seeking to gain an advantage in real-time, intelligent processing of edge sensor data. The HDK enables early adopters to utilize the unique features of the GrAI One processor. GML’s HDK is the first platform in the industry that empowers edge solution companies to leverage spatiotemporal information sparsity. As a result, OEMs, developers and engineers can build innovative solutions which can be targeted at a variety of time-critical edge-applications such as autonomous navigation, multi-modal human-machine interactions, smart healthcare, high-speed robotics and more.
GrAI One is the first processor based on the company’s unique NeuronFlow technology. Neuronflow offers a leap in performance by exploiting time-based sparsity, near memory compute, and a versatile manycore platform for signal processing and neural networks. The HDK utilizes a standard FPGA SoC board to connect to various sensor modalities including cameras, microphone arrays, radar, lidar and more. It is supported by GML’s GrAIFlow SDK, capable of both conventional program execution and machine learning inference via TensorFlow, Python and C++.
The GrAI One HDK is being demonstrated at the tinyML Summit 2020, with an autonomous steering application that could be potentially leveraged for edge devices such as robots. This application consumes only tens of milliwatts across the full SoC, allowing for battery life in top-tier edge applications and devices. The GrAI One HDK is available now for early access customers for product evaluation and application programming.
The company’s VP of Business Development Mahesh Makhijani said: “Today over 95% of compute cycles consumed at the edge for real-time AI applications are wasted, since current solutions cannot leverage time-based sparsity. GrAI One changes that.”
About GrAI Matter Labs
GML’s mission is to bring ultra-low latency, low power programmable AI processors for sensor analytics and machine learning to every device on the edge. GrAI Matter Labs has offices in Paris / France, in Eindhoven / The Netherlands and in San Jose / Silicon Valley. It is led by a team of visionary and seasoned engineers, and is backed by leading investors including iBionext, 360 Capital Partners and 3T Finance. For more information, please visit www.graimatterlabs.ai.