Leader in innovative memory solutions, Micron Technology has begun mass producing high bandwidth memory semiconductors for use in Nvidia’s newest artificial intelligence (AI), while sending its shares over four percent before the bell this week.

Named HBM3E (High Bandwidth Memory 3E), Micron promises that it will consume 30 percent less power than rival offerings and is expected to aid the towering demand for semiconductors that support generative AI applications.

Although HBM3E is said to consume less power, it does, however, enables increased performance in transferring data with a large bandwidth. Chips that use HBM3E can work efficiently especially when encountering computationally heavy tasks, namely complex graphic rendering in AI applications, for instance.

The chip is said to be used for Nvidia’s next-generation H200 graphic processing units, which is due for shipment in the coming quarter and is believed to succeed the H100 chip which is known for spiking an enormous increase in revenue at the chip maker.

Tweaked with new features added with enhanced performance better than older versions, HBM3E is the latest high-speed memory technology.

This aspect makes it a favorable choice among companies like Nvidia to use in their latest products that require high performance and good power efficiency.

Primarily led by Nvidia’s supplier SK Hynix, the demand for high-bandwidth memory (HBM) chips in AI also happens to have raised the investor hopes in which Micron believes would be able to make a gradual repossession in its other markets.

HBM is seen as Micron’s highly valued products with an increased profitability thanks to the technological framework built within.

Micron views its involvement with Nvidia’s latest AI chips positioned it at the front seat of the industry.

Earlier, the innovative memory solutions provider said that it looks to $ 700 million HBM revenue this year and progress through 2025.