Intel Lunar Lake NPU Brings 48 TOPS of AI Acceleration – Uplaza

On June 4 at Computex, held in Taiwan, Intel introduced the subsequent era of its AI PC merchandise: the Lunar Lake shopper processor structure. Different reveals from Intel on the worldwide pc expo had been:

  • Pricing for the Intel Gaudi 2 and Gaudi 3 AI accelerator kits.
  • The discharge of Intel Xeon 6 processors with Environment friendly-cores (E-cores).

Intel merchandise can be found globally from the company and its distributors.

Lunar Lake processor runs at as much as 48 TOPS

Intel revealed the small print of the Lunar Lake processor, which is able to allow AI PC efficiency. Lunar Lake brings:

  • 40% decrease system-on-chip energy and greater than thrice the AI compute in comparison with the earlier era.
  • An NPU with as much as 48 trillion operations per second.
  • A brand new GPU design with Xe2 GPU cores for improved graphics efficiency and Xe Matrix Extension arrays, which is a second AI accelerator with as much as 67 TOPS.

Lunar Lake will seem in AI PCs from greater than 20 manufacturers, together with Microsoft, all through 2024.

The Lunar Lake processor can carry out 48 TOPS of AI operations. Picture: Intel Company

Opponents to Lunar Lake

Intel’s Lunar Lake competes with Qualcomm’s Snapdragon X Elite, AMD’s AI 300 collection, Apple’s M4 and an more and more crowded subject of chips designed to make generative AI work on PCs.

Corporations are competing for larger TOPS velocity and diminished energy draw to allow AI PC capabilities like Microsoft’s Copilot. AMD claims its AI 300 collection reaches 50 TOPS. NVIDIA is speaking about getting between 200 and 1,300 TOPS with its GeForce RTX GPUs for gaming and heavy-duty inventive work on PCs and workstations.

SEE: What’s generative AI, and what does it imply for enterprise?

Final week, it was introduced that Intel is a part of a promoter group for the standard communications accelerator meant for AI chips in knowledge facilities, too.

For enterprise use circumstances, quicker AI efficiency might energy seamless generative AI assistants. Copilot, as an example, runs on OpenAI’s know-how. OpenAI and Microsoft are engaged on making its generative AI sound extra like people and bear in mind the place you might need dropped a misplaced merchandise in your Copilot+ PC.

Intel Gaudi 2 and three pricing revealed

An ordinary AI equipment made up of eight Gaudi 2 AI accelerators and a common baseboard will likely be accessible to system suppliers at $65,000. Gaudi 2 is transport now.

A equipment made up of eight Gaudi 3 AI accelerators and a common baseboard will likely be accessible to system suppliers at $125,000. It will likely be accessible in Q2 2024.

Along with revealing the worth of the kits, Intel introduced at Computex that six new suppliers will likely be working with Gaudi 3. Asus, Foxconn, Gigabyte, Inventec, Quanta and Wistron have been added to the preliminary offers with Dell, Hewlett Packard Enterprise, Lenovo and Supermicro.

The Intel Xeon 6 P-Core processor is accessible now

Initially launched on the Intel Imaginative and prescient convention in April, the primary of the 2 new Xeon 6 processors was made accessible starting June 3, Intel introduced at Computex. That processor, the Intel Xeon 6 P-core (code-named Sierra Forest), is appropriate for AI, different high-performance computing tasks and cloud-native purposes.

One tier up from the Xeon 6 P-core is the Xeon 6 E-core (code-named Granite Rapids), which has a shared software program stack with the P-core model, plus:

  • Increased core density.
  • Higher efficiency per watt.
  • Decrease power prices.

The Xeon 6 E-core is anticipated to ship within the third quarter of 2024.

A employee on the Intel Kulim Take a look at Meeting Take a look at facility in Malaysia examines Intel Xeon 6 processors with E-cores. Picture: Intel Company

“AI is driving one of the most consequential eras of innovation the industry has ever seen,” stated Intel CEO Pat Gelsinger in a press launch. “The magic of silicon is once again enabling exponential advancements in computing that will push the boundaries of human potential and power the global economy for years to come.”

Share This Article
Leave a comment

Leave a Reply

Your email address will not be published. Required fields are marked *

Exit mobile version