
DEEPX selects Samsung 2nm process for DX-M2 edge AI chip
The AMW Read
The foundry process adds detail to an already reported chip announcement, with potential impact concentrated in edge inference hardware.
DEEPX selects Samsung 2nm process for DX-M2 edge AI chip
DEEPX says its DX-M2 AI chip will use Samsung Electronics Foundry's SF2P 2nm process and target large language and vision-language model inference at about 5 watts. The company said it completed tape-out in early November. It expects first silicon in March 2027, customer evaluation samples in the second quarter, initial production shipments in the fourth quarter, and full-scale production in 2028. The chip also received a CES 2027 Innovation Award in the AI category.
The design targets robots and industrial equipment that run generative AI on the device without a cloud connection. That puts DEEPX in the edge inference market, where power consumption and local processing can shape hardware choices. The Samsung foundry process is a concrete addition to DEEPX's previously reported low-power roadmap. However, the DX-M2's performance and power claims remain to be verified after samples become available.
For builders evaluating edge hardware, the next decision point is the 2027 sample program: model performance and power consumption on their own workloads will matter alongside the production schedule. Investors can also watch whether DEEPX converts its first-generation traction into DX-M2 adoption. The company said its DX-M1 received 88 commercial purchase orders worth $13.3 million across more than 10 countries through October. Those orders concern the earlier chip and do not establish demand for DX-M2.


