MemryX Expands Cascade Platform with New Edge AI Accelerators

Jun 23, 2026
MemryX has introduced three new products in its Cascade platform family, built on the MX3 accelerator architecture, to support AI deployment from embedded systems to edge servers.

MemryX has expanded its Cascade deployment platform with three new products built on its MX3 accelerator architecture, announced in a press release. The new accelerators aim to support AI deployment across embedded systems, edge devices, and high-density edge server infrastructure. The products were introduced at Automate 2026 in Chicago.

The Cascade 100P PCIe Accelerator is designed for edge server environments and multi-camera AI applications. It features a 16-chip MX3 array that supports parallel inference on hundreds of video streams. The single-slot, passively cooled card integrates into standard servers to deliver high compute density without additional thermal or power constraints.

The Cascade 100U USB Accelerator connects through USB Type-C and includes two MX3 processors. It allows developers to add AI acceleration to existing systems or quickly prototype AI vision applications without redesigning hardware.

The Cascade 100R Raspberry Pi HAT+ provides hardware-accelerated AI inference for the Raspberry Pi ecosystem. It uses two MX3 processors connected through PCIe and can be cascaded for larger deployments. All three devices are supported by the MemryX Developer Hub, which provides a unified software environment for compiling and deploying AI models across MX3 products.

We hope you enjoyed this article

Consider subscribing to one of our newsletters like Industrial AI Weekly or Daily AI Brief.

Also, consider following us on social media:

Free newsletter

Industrial AI Weekly

The latest advancements in smart manufacturing, predictive maintenance, AI-driven quality control, supply chain & more.

Whitepaper

Tensordyne Napier: What If One Rack Could Do the Work of Nine?

Tensordyne

This Tensordyne whitepaper presents Napier, an inference-focused AI processor and rack-scale system based on the company’s TDN Math logarithmic number system. It examines infrastructure requirements for large mixture-of-experts and agentic models, compares major inference architecture approaches, and details the TDN AIP processor, TDN72 pod, TDN Link fabric, and Napier Ultra configuration. The paper reports simulation-based performance, cost, and accuracy-validation results, including Tensordyne’s projected comparison of one Napier rack with a nine-rack Nvidia Rubin plus Groq deployment; the chip is reported as taped out and in fabrication.

Read more
Free, six days a week

Daily AI Brief: the AI news that matters, in your inbox.