BrainChip Holdings Ltd. announced the AKD1500 PCIe development card, an ultra-low power PCIe card with self-learning capability that delivers silicon evaluation to any desktop or industrial PC.
Built on BrainChip’s AKD1500 Edge AI Co-Processor, the AKD1500 PCIe card installs in a standard PCIe slot in a desktop, workstation, industrial PC or single-board computer, and runs models built in existing frameworks. It is available for purchase now through BrainChip’s online store.
The launch means BrainChip’s AKD1500 chip is available in every form an engineering team needs to build and test edge AI applications: an evaluation card, a deployable M.2 module, packaged/unpackaged silicon and intellectual property (IP) a chip team can license.
It also means Akida can sit in a server anywhere and be tested as part of a team’s development pipeline. Any developer with a PC can now validate Akida with their own models using their own streaming data; a model validated on the card runs on the deployed hardware without change.
“Our customers no longer have to choose between proving out neuromorphic AI and productizing it,” said Sean Hehir, CEO of BrainChip. “Akida is available in every form they need — from card to IP —turning a technology into a complete roadmap for adoption.”
Universal Access Is the BrainChip Advantage
From a low-cost PCIe developer card to M.2 modules, NICLA/Arduino boards, silicon, die and Akida IP cores for custom ASICs, BrainChip’s complete product line puts neuromorphic computing within reach at every stage.
“The first challenge in evaluating new AI architectures is getting your hands on silicon and tools,” said Steve Brightfield, chief product officer of BrainChip. “If you have a PC with a spare slot, you can be running your own models on Akida this afternoon — and what you execute on that card seamlessly carries through to a module, a chip or IP in your own SoC design. The line of adoption is complete.”
BrainChip’s completion of universal access is well-timed given the increase in the number of edge devices incorporating edge AI silicon, such as industrial sensors, cameras, robots and wearables.
ABi Research’s 2Q 2026 market update estimates the edge AI chipset market at $34.4 billion in 2026, growing to $96 billion by 2031, with unit shipments more than doubling from 711 million to 1.59 billion.
Edge devices demand intelligence that runs on a fraction of a watt without a fan or a cloud connection, and they must keep learning once deployed. That is precisely the class of design for which Akida was built; now every team building one can validate it themselves.
The AKD1500 PCIe card launch builds on BrainChip’s industry-leading momentum, which includes production of the AKD1500 chip and launch of the M.2 module over the past two months.
