Absolute Performance at the Edge
Built around a patented Polymorphic Dataflow Architecture, supported by a comprehensive SDK, Kinara Ara Edge AI processors accelerate and optimize real-time decision making for unrivaled edge AI solutions. Our Ara AI accelerators power smart edge devices and gateways that demand responsive AI computing with optimal energy efficiency.
Uncompromised Performance with a Low Energy Footprint in a Small Form Factor
Ara Edge AI processors and modules deliver scalable performance options to support applications with stringent power demands or the highest computing requirements.
Ara-1 Edge AI Processor
From cameras and sensors through edge gateway servers, Ara-1 Edge AI Processors deliver the ideal balance of computing performance and power efficiency to optimize intelligent applications at the edge.
Latency Optimized Design
Optimized to deliver results for a batch size of ‘1’, the Ara-1 Edge AI processor ensures real-time responsiveness to meet the needs of deterministic edge applications.
Bigger, Better Deep Learning Applications
Ara Edge AI Processors can effectively run multiple models without incurring any switch-time performance penalties, generating low latency results with higher accuracy.
Zero Load on Host
Unique design ensures that compiled models place no load on host processors, freeing them from inference application demand.
A New Standard in Flexibility
The processor provides incredible flexibility in implementing highly scalable, low-latency, low power applications.
KU-1 USB Module
KM-1 M.2 (M-key) Module
Edge AI Appliance
Discover unrivaled power and efficiency with our state-of-the-art edge AI appliance, EAIA, manufactured and sold by VVDN Technologies. The appliance is uniquely engineered for real-time video analytics on 8 parallel streams, each at a resolution of 1080p30. Powered by AMD Kria™ K26 SOM and 4 Kinara Ara-1 edge AI processors, EAIA redefines the boundaries of video processing and enables multiple models to be run on each stream at real time.
U.2 Module
For edge servers that demand the highest performance at the lowest latency and power, Kinara’s U.2 module scales in performance, carrying up to 4 discrete Kinara Ara-1 Edge AI Processors on a single module and delivering the best performance suitable for large AI workloads.