About INTERA

INTERA is an Edge Computing Systems Company focused on advanced semiconductor IP for high-performance, energy-efficient AI processing at the edge. Our core business is the development and licensing of scalable AI computing architectures and acceleration engines including Neural Processing Units (NPUs), dedicated AI accelerators, and neuromorphic hardware blocks for seamless integration into SoCs and embedded devices. With a strong focus on TinyML and embedded AI, our technology enables real-time, on-device inference in resource-constrained, battery-powered, and always-on systems. INTERA solutions combine optimized hardware IP with supporting software stacks compilers, SDKs, runtime frameworks, and model-optimization toolchains to deliver superior computing performance, ultra-low power consumption, reduced memory footprint, smaller silicon area, lower latency, and high integration flexibility. Our technologies address applications across manufacturing, energy, healthcare, logistics, IoT, robotics, smart infrastructure, and defense.

At INTERA we are aware that hardware design has become a core enabler of innovation for the age of AI, as it is presenting a unique set of challenges to its pioneers, both at cloud and edge segments.

INTERA’s AI design portfolio for edge devices focuses on improving the quality-of-results (QoR) and time-to-results (TTR) for integrated circuit (IC) design, offering robust intellectual property (IP) solutions that enhance efficiency and performance in the design process.

Our AI Accelerator design considers:

Power Efficiency
By designing accelerators that focus on power efficiency and thermal management, clients can effectively deploy AI and compute-intensive workloads on EDGE devices.

To save power, our AI accelerators employ RPA (reduced precision arithmetic), making Neural networks still highly functional at 16-bit or even 8-bit floating point numbers, compared to the 32 bits general-purpose chips. This means they can achieve faster processing speeds at lower energy expenditure without sacrificing accuracy.

This targeted approach can unlock new applications for edge computing, such as real-time video analytics, robotics, and autonomous systems, where fast processing and low latency are critical.

Low latency
Our accelerators improve the way data is moved from one place to another, for INTERA this is critical when trying to optimize AI workloads. 

Our AI accelerators use different memory architectures, allowing them to achieve lower latencies and better throughput. These specialized design features, including on-chip caches and high-bandwidth memory, are vital to speeding the processing of large datasets necessary for high-performance.

High performance
Our design ensures efficient processing without compromising device longevity or functionality.

By implementing robust and efficient systems, our accelerators can effectively support a wide range of applications and environments, such as IoT devices, edge computing platforms, and mobile systems, all these often require scalable, reliable, and low-latency processing.