Intel unveils its Lunar Lake chip for AI PCs

The new chip will deliver more than three times the AI compute and is set to be released in September.

Last month, Microsoft and other major laptop vendors announced Copilot+ PCs powered by the latest chip from Qualcomm, the Snapdragon Elite X.

On Monday, AMD unveiled its Ryzen series, which will power the next wave of AI PCs. And soon, Intel will bring its own hardware to the game.

The company detailed its flagship processor Lunar Lake at the Computex conference in Taipei. It’s set to power more than 80 different AI PC designs from 20 vendors.

According to Intel, Lunar Lake will deliver up to 40% lower system on-chip power and more than three times the AI compute.

Lunar Lake will be equipped with Intel’s fourth generation neural processing unit (NPU), which can deliver up to 48 tera-operations per second (TOPS) of AI performance. It will meet Microsoft’s requirements for Copilot+ PC, which are 40 TOPS.

The NPU, whose main purpose is to take on AI tasks, would be nearly four times more than the previous version chip called Meteor Lake.

Intel’s Lunar Lake chips will feature a new design and combine Xe2 GPU cores for graphics and Xe Matrix Extension (XMX) arrays for AI.

According to Intel, the Xe2 GPU cores improve gaming and graphics performance by 1.5x over the previous generation, while the new XMX arrays enable a second AI accelerator with up to 67 TOPS of performance for extraordinary throughput in AI content creation.

Lunar Lake will also feature an advanced low-power island, a compute cluster that handles background and productivity tasks more efficiently, resulting in better battery life.

Intel claims that the upgraded connectivity suite integrates WiFi 7.0, Bluetooth 5.4, PCIe Gen5 and PCIe Gen4 ports, and Thunderbolt™ 4 ports.

The company expects its latest processors to be shipped in the third quarter of the year.

Intel has also unveiled its latest processors for AI training, aiming to compete with Nvidia and AMD.

The new platform will be called Xeon 6, with Xeon 6 E-core available on the day of announcement, while Xeon 6 P-cores are expected to launch next quarter.

Intel says that the chip will address the broad array of use cases and workloads, from AI and other high-performance compute needs to scalable cloud-native applications.