BrainChip Holdings Ltd, a global pioneer in ultra-low power, fully digital, event-based neuromorphic AI, has unveiled its latest innovation the AKD1500 neuromorphic Edge AI accelerator co-processor chip at Embedded World North America.
With this new release, BrainChip aims to revolutionize Edge AI by combining power efficiency and high performance. The AKD1500 delivers an impressive 800 giga operations per second (GOPS) while consuming less than 300 milliwatts, setting a new industry benchmark for energy-efficient AI processing. Consequently, this chip becomes a perfect fit for battery-powered wearables, smart sensors, and heat-constrained devices, where energy efficiency and thermal management are crucial.
AI Authority Trend: BrainChip Partners with RTX’s Raytheon for AFRL Radar Contract
Moreover, the AKD1500 seamlessly integrates with x86, ARM, and RISC-V host processing platforms through PCIe or Serial interfaces, making it easy for developers to adopt across varied applications. This co-processor architecture is designed to upgrade existing multi-processor SoCs in defense, industrial, and enterprise environments while also enhancing embedded microcontrollers for AI-driven healthcare, consumer electronics, and wearable solutions without requiring a complete system redesign. Notably, the AKD1500 has already been deployed in several AI-enabled sensing solutions for medical and defense use cases, including partnerships with Parsons, Bascom Hunter, and Onsor Technologies.
“The AKD1500 is a catalyst for the next wave of intelligent AIoT devices,” said Sean Hehir, CEO of BrainChip. “We’re empowering developers to break free from cloud dependency and bring adaptive learning directly to the edge in a compact, cost-effective package. This technology will make AI truly ubiquitous in smart factories, homes, and wearable devices.”
Adding to this, Anand Rangarajan, Director of AI & IOT Compute at GlobalFoundries, emphasized the chip’s design excellence. “BrainChip’s AKD1500 on our 22FDX platform delivers outstanding compute and memory efficiency,” he said. “Using BrainChip’s neuromorphic architecture combined with GlobalFoundries’ 22FDX process technology, the AKD1500 offers an excellent performance, power and cost envelope that fits into edge devices. We’re proud to support BrainChip’s end-to-end embedded AI solutions using GlobalFoundries silicon.”
AI Authority Trend: BrainChip and Onsor Partner to Develop AI-Powered Seizure-Detecting Glasses
To further accelerate development, BrainChip supports the AKD1500 with its MetaTF software development environment, allowing engineers to convert, quantize, compile, and deploy AI models seamlessly from standard TensorFlow/KERAS formats. This not only reduces development time and cost but also broadens access for AI developers seeking to implement adaptive intelligence at the edge.
In addition, BrainChip’s event-based Akida neuromorphic architecture enables on-chip learning a key differentiator that eliminates the need for cloud-only training, setting it apart from traditional AI accelerators.
AKD1500 samples are already available, with mass production slated for Q3 2026. Meanwhile, Jonathan Tapson, BrainChip’s Chief Development Officer, will present “The Impact of GenAI Workloads on Compute-in-Memory Architectures” at Embedded World North America on November 4th, further reinforcing the company’s thought leadership in neuromorphic AI innovation.
AI Authority Trend: BrainChip Provides Low-Power Neuromorphic Processing for Quantum Ventura’s Cyberthreat Tool
To share your insights, please write to us at info@intentamplify.com
