BrainChip Launches AKD1500 Edge AI Co-Processor

BrainChip Holdings Ltd (ASX: BRN, OTCQX: BRCHF, ADR: BCHPY), a world chief in ultra-low energy, totally digital, event-based neuromorphic AI, at this time introduced the launch of its AKD1500, a neuromorphic Edge AI accelerator co-processor chip, at Embedded World North America.

Designed to ship distinctive efficiency with minimal energy consumption, the AKD1500 achieves 800 giga operations per second (GOPS) whereas working beneath 300 milliwatts—setting a brand new benchmark for edge AI effectivity. This makes AKD1500 preferrred for deployment in battery powered wearables, good sensors, and heat-constrained environments the place battery life and thermal limits are crucial.

The AKD1500 integrates seamlessly with x86, ARM, and RISC-V host processing platforms by way of PCIe or Serial interfaces, enabling fast adoption throughout a variety of functions. The AKD1500 co-processor strategy is right for a variety of environments and industries, upgrading multi-processor SoCs inside protection, industrial and enterprise settings, and upgrading embedded microcontrollers for AI options in healthcare, wearables, and shopper electronics with out a full system redesign. The AKD1500 product has been delivered and designed into a number of finish options in AI enabled sensing for medical and protection associated functions, together with Parsons, Bascom Hunter and Onsor Technologies.

“The AKD1500 is a catalyst for the following wave of clever AIoT gadgets,” stated Sean Hehir, CEO of BrainChip. “We’re empowering builders to interrupt free from cloud dependency and produce adaptive studying on to the sting in a compact, cost-effective bundle. This expertise will make AI really ubiquitous in good factories, houses, and wearable gadgets.”

“BrainChip’s AKD1500 on our 22FDX® platform delivers excellent compute and reminiscence effectivity,” stated Anand Rangarajan, Director of AI & IOT Compute at GlobalFoundries. “Embedded builders are continually innovating to get the proper stage of AI to suit inside efficiency, energy and space constraints. Using BrainChip’s neuromorphic structure mixed with GlobalFoundries’ 22FDX® course of expertise, the AKD1500 presents a wonderful efficiency, energy and price envelope that matches into edge gadgets. We’re proud to help BrainChip’s end-to-end embedded AI options utilizing GlobalFoundries silicon.”

AKD1500 is supported by BrainChip’s MetaTF software program improvement instruments atmosphere, enabling machine studying engineers to simply convert, quantize, compile and deploy fashions on Akida utilizing normal TensorFlow/KERAS codecs which dramatically reduces improvement time and price whereas increasing accessibility for AI builders. BrainChip’s event-based Akida neuromorphic structure additionally permits the AKD1500 to offer on-chip studying, a crucial differentiator from typical AI accelerators that rely solely on cloud-based coaching.

AKD1500 samples can be found at this time with quantity manufacturing scheduled for Q3’26.

BrainChip’s Chief Development Officer, Jonathan Tapson, will current “The Impact of GenAI Workloads on Compute-in-Memory Architectures” at Embedded World North America on November 4th.

For extra info: be part of us for a demo of AKD1500 at Booth 3080; go to the BrainChip developer website free of charge tutorials, instruments, and fashions for neuromorphic computing; and take a look at the Embedded World North America website.

The publish BrainChip Launches AKD1500 Edge AI Co-Processor first appeared on AI-Tech Park.