Register for free to
view the full innovation profile.
The Stillwater Knowledge Processing Unit solves the problem of creating cost-effective hardware acceleration for embedded intelligence applications for power constrained env
About
Power efficiency and low-latency are key differentiating attributes for embedded intelligence applications. The Stillwater KPU is a distributed data flow processor capable of 100GOPS/Watt, that is, several orders of magnitude better than CPUs and GPUs for AI use cases, enabling new applications and new capabilities.
Key Benefits
Stillwater KPU: 100.0 GOPS/Watt NVIDIA GPU : 10.0 GOPS/Watt Intel CPU : 0.1 GOPS/Watt This performance difference translates into 10-100x cost reductions, or 10-100x performance improvements.
Applications
IoT, IIoT, robotics, autonomous vehicles, ADAS, AI cloud services, control systems