Intel Meteor Lake brings remarkable breakthrough in generative AI processing

Abraham

Intel demonstrated the performance of the vision processing unit (VPU, its acronym in English) of its advanced chipset model Meteor Lake, a component that promises to offer a great boost to applications of generative artificial intelligence and stable diffusion, highlighting the advances of the developer in the AI segment.

This is the company’s first processor to use disaggregated architecture technology, a feature that combines multiple building blocks onto a single chip using the Forveros 3D packaging tool.

Meteor Lake has three main types of blocks: the I/O block, the SoC block, and the compute block. The compute block includes the CPU block and the GPU block. The CPU block uses a hybrid design, with high-performance cores based on the Ocean Cove microarchitecture and high-efficiency cores based on the Crestmont microarchitecture.

[mb_related_posts1]

Reportedly, the GPU block uses an integrated version of the Arc Alchemist architecture, which supports advanced features like DX12 Ultimate, Ray Tracing, and XeSS.

The VPU block, in turn, is a dedicated component to run low-power artificial intelligence tasks on all Meteor Lake models. Intel showed two demos of the VPU in action: one generating realistic audio content from text and another generating photorealistic images from sketches.

Intel AI Meteor Lake. | Image: Playback.

According to Intel, the VPU can accelerate these types of applications by up to 10 times compared to a CPU or GPU only solution. Intel also announced that Meteor Lake is in the final stages of development and is expected to be released in the second half of 2023.

The processor will be manufactured using the Intel 4 process, which offers a 20% improvement in performance per watt over the Intel 7 process. Meteor Lake will be compatible with the new LGA 1700 socket and 600 series motherboards.

[mb_related_posts2]

What are your expectations for the generative AI segment? Tell us, comment!