BrainChip Launches Raspberry Pi, Intel-Based Development Kits for Its Akida Neuromorphic Processor

You'll be paying nearly $5,000 for the entry-level Raspberry Pi-based development kit if you want to play with an Akida.

Neuromorphic computing specialist BrainChip has announced the launch of two development kits for its Akida AKD1000 neural network processor, offering developers the choice of driving its device from a Raspberry Pi or a compact Shuttle x86 PC.

"Offering development kits is not only a major step towards full commercialization," says Anil Mankar, BrainChip co-founder and chief development officer, of the launch. "It's also an exciting opportunity to see how our partners and future customers will put Akida to work in environments and scenarios like consumer electronics, industrial applications, aerospace and defense systems, healthcare and medical devices, automotive technology, and more."

"We believe the AKD1000 silicon, or the licensing of Akida in a configurable IP format, will lead to major changes in industries using AI at the edge because of its performance, security, low power requirements, and mainly Akida’s ability to perform AI training and learning on the device itself, without dependency on the cloud."

BrainChip unveiled the Akida processor, the design of which is inspired by the neurons of the human brain, at the Linley Fall Processor Conference two years ago. The chip is claimed to offer high performance for spiking neural networks, while drawing power only when actively processing spikes — making them ideal, the company says, for edge AI tasks.

Two years on, the company is ready to prove the concept. The company has launched two development kits, the first being built around a Raspberry Pi Compute Module 4 on the standard open-hardware IO Board carrier. To this, the company has added a custom case and the real meat: A PCI Express add-in board with the Akida AKD1000 processor on-board.

The company claims its Akida processors can offer energy-efficient yet high-performance compute for spiking neural networks. (📹: BrainChip)

While ideal for those targeting embedded and low-power development, BrainChip has an alternative for those looking for a less resource-constrained target: A compact Shuttle PC built around a 10th-generation Intel Core processor and featuring the same PCIe Akida card. In both cases, buyers receive example Akida models and the Meta TF software development environment.

Those interested in trying the Akida out, however, will need to dig deep: The Raspberry Pi-based kit is available to order for $4,995, while the Shuttle version is priced at $9,995. BrainChip has not yet confirmed when ordered hardware will ship.

Gareth Halfacree
Freelance journalist, technical author, hacker, tinkerer, erstwhile sysadmin. For hire: freelance@halfacree.co.uk.
Latest articles
Sponsored articles
Related articles
Latest articles
Read more
Related articles