MSI's new Nvidia RTX Spark Mini PC rivals Mac Studio, Asus ProArt Mini PC, HP OmniDesk Mini PC

[Redaktion]

MSI has introduced the EdgeMesa N AI+ mini PC, its direct rival to the new HP OmniDesk Mini PC and Asus ProArt Mini PC. Featuring the new NVIDIA RTX Spark architecture, this compact desktop is optimized for local generative AI, LLM development, and high-performance inferencing, offering 10GbE connectivity and multi-display support.

https://www.notebookcheck.net/MSI-s-new-Nvidia-RTX-Spark-Mini-PC-rivals-Mac-Studio-Asus-ProArt-Mini-PC-HP-OmniDesk-Mini-PC.1312386.0.html

[Prassel]

look at these Mini-PCs (including the ones from Asus, HP, etc.), with no ports at the front, and very few ports at the back, they are clearly not meant as an everday device for average user.

This should be said more clearly

[Greed]

Headline should read:

MSI's new Nvidia RTX Spark Mini PC's bold new design rivals a 2003 external DVD burner.

Fixed it for you, you're welcome!

[wrong, not yet a rival]

Not a rival: A NVIDIA Spark is based on a NVIDIA's GB10 chip (273 GB/s) (en.wikipedia.org/wiki/Blackwell_(microarchitecture)#Blackwell_dies), while a Mac Studio is/was based on Apple's M4 Max (up to 546 GB/s) or M3 Ultra (819.3 GB/s) chips.

Unfortunately Apple doesn't sell higher RAM configs anymore:
en.wikipedia.org/wiki/Mac_Studio: "Configurable up to 512 GB until March 4, 2026;[30] 256 GB until May 5, 2026[31]".

If the Spark had 192 GB or 256 GB RAM configs, then even with the lower GB/s one might call it a rival, because at least then it would have more RAM than a Mac Studio.

96 GB @ 819.3 GB/s vs 128 GB @ 273 GB/s
Total memory size x speed value:
M3 Ultra: 78653 = 819.3 GB/s * 96 GB
GB10/Spark: 34944 = 273 GB/s * 128 GB

Offering only 96 GB is crazy low tho, and a waste of the huge 1024 bits APU potential (exactly like giving a RTX 5090 6 GB of VRAM (5.33 times lower: 512 GB -> 96 GB = 32 GB -> 6 GB), so low in fact, that it's going to push some users to get AMD Strix Halo or NVIDIA Spark, depending on their use case, especially considering the price.

Problem is, 96 GB @ 819.3 GB/s can fit and run dense models that outperform MoE models on the slower 128 GB RAM ;) and compensate hugely for the fact that it's not 128 GB RAM. If Qwen releases a dense e.g. 50B to 90B model, then the 819.3 GB/s start to shine even more. But we don't have to wait: Qwen3.6-27B dense is a good example.

[Bandwidth expert]

It's not 273 but 300 as they're using slightly faster memory than what's used in the DGX Spark.

run dense models that outperform MoE models

Still not providing an actual use case on how any of these models make an improvement to anyone's life.

Is the purpose just to increase energy bills?