Revolutionary 2nm AI Chip Built by Samsung Foundry Delivers Maximum Efficiency on Just 5 Watts

Samsung Foundry is successfully turning the page on its previous manufacturing challenges, securing a major victory in the contract chip manufacturing sector. Beyond refining its fabrication technologies, the division is welcoming innovative clients and proving its capability to deliver ultra-efficient silicon for next-generation artificial intelligence applications.

• ✨ DEEPX unveiled the DX-M2, an advanced on-device AI inference chip built entirely on Samsung Foundry’s cutting-edge 2nm GAA process node.
• ✨ Engineered specifically for industrial automation, smart machinery, and robotics, the processor remarkably operates on just 5 watts of power.
• ✨ The exceptional energy efficiency and design prowess recently earned the hardware a prestigious CES 2027 Innovation Award in the AI category.
• ✨ Initial tape-out is already complete, with pilot mass production scheduled for late 2027 and full-scale commercial manufacturing planned for 2028.

DEEPX Introduces the Ultra-Efficient DX-M2 Powered by Samsung Technology

DEEPX, a prominent South Korean fabless AI chip firm, has officially introduced its high-performance, next-generation processor known as the DX-M2. By utilizing Samsung Foundry's state-of-the-art 2nm Gate-All-Around (GAA) manufacturing architecture, the company has created a silicon solution tailored directly for autonomous industrial devices and complex robotics.

The chip's ability to maintain high processing speeds while consuming an estimated 5 watts of electricity highlights a massive leap forward in edge computing. Because of this breakthrough power profile, the DX-M2 managed to secure a coveted CES 2027 Innovation Award within the artificial intelligence sector, validating the strength of this collaborative partnership.

Furthermore, the DX-M2 is purposefully designed to execute complex on-device inference tasks for both Large Language Models (LLMs) and Vision Language Models (VLMs). This localized capability allows autonomous robots and smart industrial systems to execute advanced machine learning routines completely independently, eliminating any dependency on cloud infrastructure or external data centers.

What makes the new DEEPX DX-M2 chip unique?

The DX-M2 stands out due to its combination of Samsung's advanced 2nm GAA manufacturing node and an ultra-low power consumption profile of just 5 watts.

What applications are targeted by this 2nm processor?

The chip is specifically engineered for robotics, industrial machinery, and advanced on-device edge computing scenarios requiring local AI inference.

Does the chip rely on cloud connectivity to run artificial intelligence models?

No, the DX-M2 handles inference for Large Language Models and Vision Language Models locally on the device without needing external data centers.

When will the mass production of the DX-M2 begin?

Following a successful tape-out, initial mass production is slated for the final quarter of 2027, with full-scale production expanding into 2028.

🔎 As Samsung Foundry continues to scale up its advanced manufacturing processes and secure major design wins, innovations like the DEEPX DX-M2 demonstrate a promising horizon for energy-efficient semiconductor engineering and autonomous edge hardware.