About MemryX
MemryX makes the MX3, an accelerator chip for running trained AI models at the edge, meaning in devices rather than data centres. Its at-memory design places processing elements next to the memory that holds the model weights, which reduces the energy spent moving data, and it uses a dataflow approach where results pass directly between compute units. The company sells an M.2 module carrying four MX3 chips rated at up to 24 TOPS, or trillions of operations per second. MemryX was formed in 2019 at the University of Michigan by professors Wei Lu and Zhengya Zhang and is based in Ann Arbor.
MemryX was founded in 2019 and is based in Ann Arbor, MI, United States. It is listed under Memristors & ReRAM, AI Accelerators. Reported funding is $44M Series B (March 2025).
Compute Technologies
- Memristors & ReRAM
- AI Accelerators
What MemryX works on
Memristors & ReRAM
A memristor is an electronic component whose resistance depends on the current that has passed through it, and it keeps that resistance when power is off. Leon Chua described the idea in 1971. In 2008 researchers at HP Labs published a device they linked to it.
AI Accelerators
An AI accelerator is a chip, or a part of a chip, designed to run machine learning faster or with less energy than a general-purpose processor. The category includes GPUs, TPUs, neural processing units (NPUs) and chips from many startups.
Also in United States
Oxmiq Labs, Lemurian Labs, Crystal Group, Tesla Dojo, deepsilicon
