Samsung begins mass production of industry-leading AI chip for mobile

Seoul, Aug 6 (IANS) Samsung Electronics, the world's largest memory chip maker, said on Tuesday it has started mass production of the industry's thinnest mobile DRAM, LPDDR5X DRAM, tailored to on-device artificial intelligence.

The new 12-nanometer (nm)-class 12 gigabyte (GB) and 16GB LPDDR5X DRAM packages are only 0.65 mm high, making them the thinnest LPDDR DRAM in the industry, according to Samsung Electronics, reports Yonhap news agency.

The new chip, stacking four layers of 12nm-DRAM die, is 9 per cent thinner and 21.2 per cent more efficient in heat resistance compared to its predecessor.

Samsung said the ultra-slim LPDDR5X DRAM packages can create additional space within mobile devices, enabling high-performance applications and advanced features like on-device AI.

"Samsung's LPDDR5X DRAM sets a new standard for high-performance on-device AI solutions, offering not only superior LPDDR performance but also advanced thermal management in an ultra-compact package," said Bae Yong-cheol, executive vice president at Samsung Electronics.

The company has also set up a unit dedicated to developing high bandwidth memory (HBM) in a bid to regain the leadership in the AI semiconductor market.

This move underscores the tech giant's commitment to enhancing its R and D structure for HBM, a high-performance DRAM in high demand, particularly for Nvidia's graphics processing units, which are key for AI computing.

Samsung’s second quarter (Q2) operating profit surged more than 15 times from a year earlier on the robust performance of its semiconductor business.

The company said its chip business earned 28.6 trillion won in sales for the three-month period through June, with an operating profit of 6.45 trillion won.

The strong performance was driven by increasing demand for memory chips, including those for artificial intelligence servers and solid-state drives.

Disclaimer: This story has not been edited by the Sakshi Post team and is auto-generated from syndicated feed.