16 comments

  • anesxvito3 minutes ago
    The philosophy of knowing exactly what's on your system translates directly to how you think about software you build. Local-first, no telemetry, minimal dependencies. FreeBSD instilled that mindset in a generation of developers that now pushes back hard against cloud-everything SaaS. Tauri over Electron is the same argument applied to desktop apps.
    • brazukadev0 minutes ago
      &gt; Tauri over Electron is the same argument applied to desktop apps.<p>you lost me here but still got my upvote. Tauri and Electron are pretty much the same, compared to local-first vs cloud.
  • rka1282 minutes ago
    &quot;democratize access to AI and accelerating innovation.&quot;<p>So they make inference cheaper and the models get even worse. Or Jensen Huang has AI psychosis. Or both.<p>Here is a new business idea for Nvidia: Give me $3000 in a circular deal which I will then spend on a graphics card.
  • recvonline10 minutes ago
    Does this mean their gaming GPUs are becoming less in demand, and therefore cheaper&#x2F;more available again?
    • TheRoque8 minutes ago
      It means it will be profitable to mine crypto again
  • baal80spam25 minutes ago
    Say what you want about NVIDIA (to me they are just doing what every company would do in their place), but they create engineering marvels.
  • d_silin1 hour ago
    It is a 88-core ARM v9 chip, for somewhat more detailed spec.
    • mixmastamyk39 minutes ago
      Hmm, the 128-core Ampere Altra CPU is already available, and in a case from System76. I wonder what else differentiates it.<p>If they&#x27;re going to build CPUs I wish they had used Risc-V instead. They are using it somewhat already.
  • kibibu22 minutes ago
    Am I crazy, or is Jensen&#x27;s statement a copy-paste from ChatGPT?<p>(Could be both)
    • wmf20 minutes ago
      If AI is so great why should he not use it?
  • gcanyon35 minutes ago
    Anyone know how this compares to Apple’s M5 chips? Or is that comparison &lt;takes off sunglasses&gt; apples to oranges.
    • pdpi25 minutes ago
      Features like hardware FP8 support definitely make it apples-to-oranges.
    • storus29 minutes ago
      Grace GB10, Vera&#x27;s predecessor, had a single core performance comparable to M3 so I guess we can expect at least M4 level performance now.
      • porphyra13 minutes ago
        Isn&#x27;t the GB10 a Mediatek chip and not directly related to the Grace datacenter CPU?
    • d_silin31 minutes ago
      M5 are 9-18 cores and optimized for power-efficiency, those are more like Xeons, with 200-300W TDP, I&#x27;d bet.
      • kllrnohj21 minutes ago
        If M5 has 9-18 cores and takes ~20w, then that&#x27;s ~1-2w per CPU core. If these are 200-300W, and have ~100-200 CPU cores, then guess what? That&#x27;s also ~1-2w per CPU core.<p>Xeons, Epycs, whatever this is - they are all also typically optimized for power efficiency. That&#x27;s how they can fit so many CPU cores in 200-300W.
  • tencentshill1 hour ago
    So does this cut out Intel&#x2F;x86 from all the massive new datacenter buildouts entirely? They&#x27;ve already lost Apple as a customer and are not competitive in the consumer space. I don&#x27;t see how they can realistically grow at all with x86.
    • alecco56 minutes ago
      Even Apple hardware looks inexpensive compared to Nvidia&#x27;s huge premium. And never mind the order backlog.<p>x86 and Apple already sell CPUs with integrated memory and high bandwidth interconnects. And I bet eventually Intel&#x27;s beancounter board will wake up and allow engineering to make one, too.<p>But competition is good for the market.
      • storus28 minutes ago
        Apple went from a high-end PC to a low-end AI provider due to blocking Nvidia on their platform.
    • mikrl24 minutes ago
      &gt;are not competitive in the consumer space<p>AFAIK they still dominate on clock rate, which I was surprised to see when doing some back of the envelope calculations regarding core counts.<p>I felt my 8 core i9 9900K was inadequate, so shopped around for something AMD, and IIRC the core multiplier of the chip I found was dominated by the clock rate multiplier so it’s possible that at full utilization my i9 is still towards the best I can get at the price.<p>Not sure if I’m the typical consumer in this case however.
      • kllrnohj12 minutes ago
        Your 9900k at 5ghz does work slower than a Ryzen 9800X3D at 5ghz. A <i>lot</i> slower (1700 single core geekbench vs 3300, and just about any benchmark will tell the same story). Clock speed alone doesn&#x27;t mean anything.
      • wmf11 minutes ago
        A 9700X is twice the performance of a 9900K and M5 Max is almost 3X the performance. The megahertz myth is a myth.
  • rishabhaiover59 minutes ago
    I&#x27;m assuming this is for tool call and orchestration. I didn&#x27;t know we needed higher exploitable parallelism from the hardware, we had software bottlenecks (you&#x27;re not running 10,000 agents concurrently or downstream tool calls)<p>Can someone explain what is Vera CPU doing that a traditional CPU doesn&#x27;t?
    • kibibu25 minutes ago
      &gt; you&#x27;re not running 10,000 agents concurrently or downstream tool calls<p>Cursor seem to be doing exactly that though
    • urig53 minutes ago
      Lots and lots of CPUs pooled. Faster more efficient power RAM accessible to both GPU and CPU. IIUC.
      • rishabhaiover36 minutes ago
        But at what stage are we asking for that RAM? if it&#x27;s the inference stage then doesn&#x27;t that belong to the GPU&lt;&gt;Memory which has nothing to do with the CPU?<p>I did see they have the unified CPU&#x2F;GPU memory which may reduce the cost of host&#x2F;kernel transactions especially now that we&#x27;re probably lifting more and more memory with longer context tasks.
  • jauntywundrkind1 hour ago
    Given the price of these systems the ridiculously expensive network cards isn&#x27;t such a huge huge deal, but I can&#x27;t help but wonder at the absurdly amazing bandwidth hanging off Vera, the amazing brags about &quot;7x more bandwidth than pcie gen 6&quot; (amazing), but then having to go to pcie to network to chat with anyone else. It might be 800Gbe but it&#x27;s still so many hops, pcie is weighty.<p>I keep expecting we see fabric gains, see something where the host chip has a better way to talk to other host chips.<p>It&#x27;s hard to deny the advantages of central switching as something easy &amp; effective to build, but reciprocally the amazing high radix systems Google has been building have just been amazing. Microsoft Mia 200 did a gobsmacking amount of Ethernet on chip 2.8Tbps, but it&#x27;s still feels so little, like such a bare start. For reference pcie6 x16 is a bit shy of 1Tbps, vaguely ~45 ish lanes of that.<p>It will be interesting to see what other bandwidth massive workloads evolve over time. Or if this throughout era all really ends up serving AI alone. Hoping CXL or someone else slims down the overhead and latency of attachment, soon-ish.<p>Maia 200: <a href="https:&#x2F;&#x2F;www.techpowerup.com&#x2F;345639&#x2F;microsoft-introduces-its-newest-ai-accelerator-maia-200" rel="nofollow">https:&#x2F;&#x2F;www.techpowerup.com&#x2F;345639&#x2F;microsoft-introduces-its-...</a>
    • bob10291 hour ago
      &gt; It might be 800Gbe but it&#x27;s still so many hops, pcie is weighty.<p>Once you need to reach beyond L2&#x2F;L3 it is often the case that perfectly viable experiments cannot be executed in reasonable timeframes anymore. The current machine learning paradigm isn&#x27;t that latency sensitive, but there are other paradigms that can&#x27;t be parallelized in the same way and are very sensitive to latency.
    • babelfish1 hour ago
      Most of the big AI&#x2F;HPC clusters these systems are aimed at aren’t running regular PCIe Ethernet between nodes, they’re usually wired up with InfiniBand fabrics (HDR&#x2F;NDR now, XDR soon)
  • urig54 minutes ago
    What the heck is agentic inference and how is it supposed to be different from LLM inference? That&#x27;s a rhetorical question. Screw marketing and screw hype.
  • dmitrygr1 hour ago
    &gt; Purpose-Built for Agentic AI<p>From the &quot;fridge purpose-built for storing only yellow tomatoes&quot; and &quot;car only built for people whose last name contains the letter W&quot; series.<p>When can this insanity end? It is a completely normal garden-variety ARM SoC, it&#x27;ll run Linux, same as every other ARM SoC does. It is as related to &quot;Agentic $whatever&quot; as your toaster is related to it
    • pdpi28 minutes ago
      &gt; It is as related to &quot;Agentic $whatever&quot; as your toaster is related to it<p>These things have hardware FP8 support, and a 1.8TB&#x2F;s full mesh interconnect between CPUs and GPUs. We can argue about the &quot;agentic&quot; bit, but those are features that don&#x27;t really matter for any workload other than AI.
      • kibibu23 minutes ago
        Would cloud gaming platforms benefit from the interconnect?
        • pdpi19 minutes ago
          Don&#x27;t think they would. Games aren&#x27;t nearly as hungry for memory bandwidth as LLMs are. Also, I expect that the VRAM&#x2F;GPU&#x2F;CPU balance would be completely out of whack. <i>Something</i> would be twiddling its thumbs waiting for the rest of the hardware.
      • dmitrygr20 minutes ago
        mem bw between cores matters for .... literally all workloads that are not single-core (read: all). And FP8 matters not at all cause inference on cpu is too slow to be of any use whatsoever in the days of proper accelerators
    • dpe821 hour ago
      The power and importance of marketing is deeply underappreciated by us technical types.
      • dwb8 minutes ago
        I don’t underappreciate it, but I do despise it.
      • LogicFailsMe26 minutes ago
        And yet more than a little Gavin Belson &quot;Box III&quot; vibes here. Fortunately, no signature edition.
  • BoredPositron54 minutes ago
    Who wants general computing anyways?
  • KnuthIsGod39 minutes ago
    China will beat this....<p>Seems like a triumph of hype over reality.<p>China can do breathless hype just as well as Nvidia.
  • FridgeSeal32 minutes ago
    Are we rapidly careening towards a world where _only_ AI “computing” is possible?<p>Wanted to do general purpose stuff? Too bad, we watched the price of everything up, and then started producing only chips designed to run “ai” workloads.<p>Oh you wanted a local machine? Too bad, we priced you out, but you can rent time with an ai!<p>Feels like another ratchet on the “war on general purpose computing” but from a rather different direction.