Nvidia DGX Spark • Hacker News Board

Re: Nvidia DGX Spark

#81

Post by DrNosferatu » Thu, Aug 28, 2025, 3:36 AM UTC

Now we need a threeway benchmark between this DGX Spark, a maxed out AMD Strix* and the Mac 512GB.

Re: Nvidia DGX Spark

#82

Post by canucker2016 » Thu, Aug 28, 2025, 3:36 AM UTC

gardnr wrote: post #71

ls612 wrote: post #13

Is this the $3500 one?

There are cheaper ASUS and MSI versions "coming soon" with the same chip and less storage / memory.

on the reserve now page for USA, there's:

  ASUS Ascent GX10 - 1TB $2,999

  MSI EdgeXpert MS-C931 - 4TB $3,999

the 1TB/4TB seems to be the size of the included NVMe SSD.

the reserve now page also lists

  NVIDIA DGX Spark Bundle
  2 NVIDIA DGX Spark Units - 4TB with Connecting Cable $8,049

The DGX Spark specs lists an NVIDIA ConnectX-7 Smart NIC which is rated at 200Gbe to connect to another DGX Spark, for about double the amount of memory for models.

Re: Nvidia DGX Spark

#83

Post by maddynator » Thu, Aug 28, 2025, 3:51 AM UTC

So Raspberry Pi With GPU?

Re: Nvidia DGX Spark

#84

Post by rightisleft » Thu, Aug 28, 2025, 4:07 AM UTC

x2tyfi wrote: post #29

MBCook wrote: post #12

I’m not in this space, so I don’t know what’s normal, but I guess I’m a little surprised to see only 10 gig Ethernet for high speed connectivity. Yeah, it’s miles better than WiFi. But if there was something I’d think maybe benefit from Thunderbolt this would’ve been it. The ability to transfer large models or datasets that way just seems like it would be much faster and a real win for some customers.

You’re almost always going to bottleneck on your home internet or upstream ISP, rather than this local interface. That being said, you aren’t going to be waiting too long either way, depending on download speed. Deepseek R1 is 671GB. Multiply by 8 to get into bits: 5368Gb At full 10gbps (which, again, you probably won’t get): 5368Gb / 10gbps = 537 seconds to download 537s / 60 = 8.95 minutes. Call it 10m with overhea…

I think I interviewed you the other day and you didn’t get the job…

Re: Nvidia DGX Spark

#85

Post by skhameneh » Thu, Aug 28, 2025, 4:16 AM UTC

qingcharles wrote: post #77

conradev wrote: post #62

Earlier quoted context omitted.

where does an RTX Pro 6000 Blackwell fall in this? I feel like that’s the next step up in performance (and about the same price as two Sparks)

I thought the 6000 was slightly lower throughput than 5090, but obviously has a shitload more RAM.

It's more throughput, but way less value and there's still no NVLink on the 6000. Something like ~4x the price, ~20% more performance, 3x the VRAM.

There's two models that go by 6000, the RTX Pro 6000 (Blackwell) is the one that's currently relevant.

Re: Nvidia DGX Spark

#86

Post by wiredpancake » Thu, Aug 28, 2025, 4:27 AM UTC

rightisleft wrote: post #84

x2tyfi wrote: post #29

Earlier quoted context omitted.

You’re almost always going to bottleneck on your home internet or upstream ISP, rather than this local interface. That being said, you aren’t going to be waiting too long either way, depending on download speed. Deepseek R1 is 671GB. Multiply by 8 to get into bits: 5368Gb At full 10gbps (which, again, you probably won’t get): 5368Gb / 10gbps = 537 seconds to download 537s / 60 = 8.95 minutes. Call it 10m with overhea…

I think I interviewed you the other day and you didn’t get the job…

What?

Re: Nvidia DGX Spark

#87

Post by KingOfCoders » Thu, Aug 28, 2025, 4:36 AM UTC

nodesocket wrote: post #76

wmf wrote: post #74

Earlier quoted context omitted.

Ultras are pretty expensive.

I mean the spark is $3,999 and current M3 Max 28-Core CPU 60-Core GPU is the same price. I would expect the refreshed studio will stay around the same price.

In Germany the 96gb version is 5000 EUR and the 256gb version is 7000 EUR (no 128gb available as far as I can see).

Re: Nvidia DGX Spark

#88

Post by KingOfCoders » Thu, Aug 28, 2025, 4:45 AM UTC

I think it depends on your model size

   Fits into 32gb: 5090
   Fits into 64gb - 96gb: Mac Studio
   Fits into 128gb: for now 395+ $/token/s, 
     Mac Studio if you don't care about $ 
     but don't have unlimited money for Hxxx

This could be great for models that fit 128gb and you want best $/token/s (if it is faster than a 395+).

Re: Nvidia DGX Spark

#89

Post by larodi » Thu, Aug 28, 2025, 4:49 AM UTC

syntaxing wrote: post #22

While a completely different price point, I have a Jetson Orin Nano. Some people forget the kernels are more or less set in stone for product like these. I could rebuild my own Jetpack kernel but it’s not that straight forward to update something like CUDA or any other module. Unless you’re a business where your product relies on this hardware, I find it hard to buy this for consumer applications.

My experience with Jetson Nano was that it had to have its Ubuntu debloatred first (with 3rd party script) before we could get their NN something library to run the image recognition, designated to run on this device.

These seem to be highly experimental boards, even though are super powerful for their form factor.

Re: Nvidia DGX Spark

#90

Post by UncleOxidant » Thu, Aug 28, 2025, 4:51 AM UTC

cherioo wrote: post #24

The mainstream options seem to be Ryzen AI Max 395+, ~120 tops (fp8?), 128GB RAM, $1999 Nvidia DGX Spark, ~1000 tops fp4, 128GB RAM, $3999 Mac Studio max spec, ~120 tflops (fp16?), 512GB RAM, 3x bandwidth, $9499 DGX Spark appears to potentially offer the most token per second, but less useful/value as everyday pc.

> Ryzen AI Max 395+, ~120 tops (fp8?), 128GB RAM, $1999

Just got my Framework PC last week. It's easy to setup to run LLMs locally - you have to use Fedora 42, though, because it has the latest drivers. It was super easy to get qwen3-coder-30b (8 bit quant) running in LMStudio at 36 tok/sec.