Simple &
Scalable Edge AI

With MemryX, adding Edge AI Processing is as simple as adding DRAM


to Deploy

MemryX’s Edge AI solution enables 1-click compilation and performance optimization of trained AI models without accuracy drop. No hand-tuning or model retraining is required. MemryX Edge AI optimally configures for your trained AI model(s) – with 1-click.



One MemryX chip offers a lot of AI processing power. But if you need more performance or lower latency, simply linearly cascade any number of our AI Processing chips. Each die is a chiplet, and you can use the same software to process multiple models in one chip or large model(s) across multiple chips.



MemryX uses a proprietary, configurable native dataflow architecture, along with at-memory computing that sets the bar for Edge AI processing. The system architecture fundamentally eliminates the data movement bottleneck, while supporting future generations (new hardware, new processes/chemistries and new AI models) — all with the same software.

MX3 AI Accelerator



Request more information