Intel says Lunar Lake will have 100+ TOPS of AI performance — 45 TOPS from the NPU alone meets requirement for next-gen AI PCs

Intel logo
(Image credit: Tom's Hardware)

Intel CEO Pat Gelsinger flashed a Lunar Lake processor, the company's next-gen laptop chip, at its Vision 2024 event, saying the chip provides 100+ TOPS of performance in AI workloads, with 45 of those TOPS coming from the NPU alone. That's 3X the AI performance of Intel's current-gen chips and meets the bar of 45 TOPS from the NPU that the company recently discussed during its AI Summit in Taipei as the requirement for next-gen AI PCs. At that event, Intel executives, in a question-and-answer session with Tom's Hardware, also said that elements of Microsoft's Copilot will soon run locally on Windows PCs. 

Intel

(Image credit: Tom's Hardware)

Intel's current-gen Meteor Lake chips only provide 10 TOPS of performance from the NPU, falling below the bar for next-gen AI PCs. Lunar Lake's 45 TOPS from the NPU meets that bar exactly. Gelsinger didn't elaborate on how much of the remaining 55+ TOPS comes from the CPU and GPU, but it's reasonable to expect in the range of 50 TOPS from the GPU and 5 to 10 FLOPS from the CPU cores.

CPUs

Lunar Lake closeup (Image credit: Tom's Hardware)

AMD's current-gen Ryzen Hawk Point platform has an NPU with 16 TOPS of performance, which also falls below the bar for next-gen AI PCs. However, the company recently sent us a statement regarding its next-gen products: 

We believe an AI PC requires strong CPU, GPU and NPU engines, which is what AMD has been delivering for more than a year with our Ryzen 7040 and now 8040 Series. At our December Advancing AI event, we disclosed our next-gen “Strix Point” mobile processors with XDNA 2 architecture would have up to 3x the generative AI performance of the current generation. We believe this performance will position us to remain the leading choice for next-gen AI PCs." — AMD representative to Tom's Hardware.

AMD wouldn't elaborate further and break out the amount of performance delivered from each unit, but simple math puts AMD's Strix Point NPU at 48 TOPS if NPU performance triples.

Qualcomm's Snapdragon X Elite chips are the elephant in the room — these Arm chips will debut with 45 TOPS of performance from its NPU (75 TOPS total) in the "middle of the year," setting up a tight race between Intel, AMD, and Qualcomm for the lead in the next generation of AI PCs. 

Intel appears to have a running start with its Core Ultra Meteor Lake chips. Gelsinger says the company has already shipped five million AI PCs to date and plans to ship 40 million units by the end of the year. That's a solid start towards the company's goal of shipping 100 million AI PC chips by the end of 2025. 

TOPICS
Paul Alcorn
Managing Editor: News and Emerging Tech

Paul Alcorn is the Managing Editor: News and Emerging Tech for Tom's Hardware US. He also writes news and reviews on CPUs, storage, and enterprise hardware.

  • NinoPino
    Big numbers. Another wave of good machines thrown in the garbage because are not "AI PCs" ?
    Reply
  • ThomasKinsley
    System TOPS vs NPU TOPS. Didn't know corporations made a distinction and I can't help but wonder if Intel is trying to make their chip sound better than it really is.
    Reply
  • ThomasKinsley
    NinoPino said:
    Big numbers. Another wave of good machines thrown in the garbage because are not "AI PCs" ?
    If anyone wants a true AI PC then they should wait a bit. The current AI PCs are neither good enough for AI or traditional usage, and they use a lot of battery. Give it 4-5 years and we'll see AI accelerators that make the current NPUs look like a joke.
    Reply
  • usertests
    45 of those TOPS coming from the NPU alone. That's 3X the AI performance of Intel's current-gen chips
    More like 4.5X the performance. Or are they comparing to the number nobody cares about?
    Reply
  • Jagar123
    At this time I just don't care about NPU TOPS performance. It comes off like ray tracing which was announced 5 years+ ago. 5 years later and I still don't care for ray tracing. I guess it's market hype and sells products but damn am I tired of that.
    Reply
  • usertests
    NinoPino said:
    Big numbers. Another wave of good machines thrown in the garbage because are not "AI PCs" ?
    If I can get a refurbished i5-12400T PC for $100 because of Copilot hype, then bless M$.
    Reply
  • JamesJones44
    NinoPino said:
    Big numbers. Another wave of good machines thrown in the garbage because are not "AI PCs" ?
    Yes and no. Right now, yes for sure. However, as applications integrate more ML models into their applications NPUs will be beneficial. The issue IMO is there is very little way to quantify slow, ok, fast at the NPU level for consumers at the moment and so most of this is looks like noise (and is until there are more content that leverages NPU compute).
    Reply
  • Alvar "Miles" Udell
    Qualcomm's Snapdragon X Elite chips are the elephant in the room

    Not really. Windows On ARM devices is still pretty hit or miss according to if the software has a native ARM version or if it needs to be emulated, and if they're anywhere near as outrageously expensive as the current Snapdragon laptops are then they're not even going to be in the same market segment.
    Reply
  • alby13
    ThomasKinsley said:
    System TOPS vs NPU TOPS. Didn't know corporations made a distinction and I can't help but wonder if Intel is trying to make their chip sound better than it really is.
    wonder no more; yes Intel is trying to make their chip sound better than it really is. Intel is tied with their advertising side, and it's off-putting. i would never take Intel's performance information without doing research.
    Reply
  • bluvg
    usertests said:
    More like 4.5X the performance. Or are they comparing to the number nobody cares about?
    I wondered about the "3x" also. I think it's total system TOPS for MTL (I've seen estimates of ~35 TOPS) vs. LNL. All the systems today seem to schedule between CPU/GPU/NPU based on perf and power; not sure how that will change as NPU TOPS go up, but it looks like the relative importance of total TOPS vs. NPU-only TOPS varies on use case.
    Reply