Intel has shared the first details of its upcoming Meteor Lake platform, which will debut later this year with a heavy focus on client-side AI performance. AI is the hot topic in computing right now, and Intel is betting big on making sure it can deliver a capable AI experience on client devices so that it doesn’t have to rely on the cloud for all of its workloads and use cases that are starting to pop up.
Meteor Lake uses a microchip design
Source: Intel
Meteor Lake is the next generation of Intel CPUs, the first one to be based on the Intel 4 process while also making use of Intel’s Foveros packaging technology in a chiplet-style design, in which the various components of the processor are built into separate tiles (or chips). This chip design allows for advanced power management, and also means that each chip can be individually targeted for specific workloads.
One of these chipsets is, of course, the CPU, which was developed with great emphasis on energy efficiency, while retaining the hybrid architecture that we saw in the 12th and 13th generation processors. There’s also an all-new GPU, as Intel finally brings Arc graphics to an integrated architecture, including support for things like ray tracing and XeSS in a low-power design.
However, a big focus for Intel is the new Artificial Intelligence Engine, or VPU (Vision Processing Unit) built into the processor. This is what Intel hopes will drive client-side AI workloads forward, greatly reducing the computing requirements for AI inference.
Intel does everything on artificial intelligence
By now, everyone has probably heard of AI, which appears in many forms, whether it is conversational AI such as ChatGPT based on a large language model, but also image generation using tools such as Midjourney and Stable Diffusion. These workloads are usually handled in the cloud, however, which can mean increased costs for software vendors, especially as these tools become more advanced and have higher computing requirements, as well as privacy concerns for users. And with features like Windows Copilot, AI will be everywhere, so Intel doesn’t want to miss it.
This is where the VPU in the Meteor Lake platform comes in. This specialized processor is designed to handle specific AI workloads more efficiently than a general purpose CPU or even a GPU, thus enabling this type of workload to run locally on the computer. The VPU is designed for continuous AI workloads, such as applying effects such as background blur, while the GPU can be useful for AI being embedded in rendering pipelines, and the CPU can be used for lightweight single inference tasks. All of these components can also be leveraged at the same time for workloads that can take advantage of each other’s strengths.
One example given by Intel was running Stable Diffusion image generation in GIMP on the Meteor Lake platform, the computer generated a complex image based on a text description, taking advantage of the CPU, GPU, and VPU, and it took about 20 second. After that, it is also possible to run the Super Resolution scaling tool, which works exclusively on the VPU, providing a high-resolution version of the image in just a few seconds. The improved efficiency and performance of the VPU also means that effects such as background blur can work better, with improved edge detection, while still using less power.
For developers, Intel is also leveraging platforms and tools like ONNX, OpenVINO, and Windows DirectML to bring these capabilities to existing applications, including GIMP, Blender, Audacity, OBS, and Adobe applications. Intel plays on its strengths, particularly the robust ecosystem of thousands of x86 applications and platforms already built around its processors. This is a huge advantage over competitors like Qualcomm, which struggles with app compatibility on Windows.
Intel hasn’t yet shared details about the availability of Meteor Lake processors, or specific CPU and GPU specs, but it’s clear that the focus is on AI, not just for Intel, but for the entire industry.
