News for hackers. Stuff that matters.

A comment on Nvidia DGX Spark: great hardware, early days for the ecosystem.

Re:Nvidia DGX Spark: great hardware, early days for the ecos... by reenorap on Wednesday October 15, 2025 @02:04AM (#45588582)
Is 128 GB of unified memory enough? I've found that the smaller models are great as a toy but useless for anything realistic. Will 128 GB hold any model that you can do actual work with or query for answers that returns useful information?
[ Parent | on HN ]
Re:Nvidia DGX Spark: great hardware, early days for the ecos... by simonw on Wednesday October 15, 2025 @02:10AM (#45588624)
There are several 70B+ models that are genuinely useful these days.

I'm looking forward to GLM 4.6 Air - I expect that one should be pretty excellent, based on experiments with a quantized version of its predecessor on my Mac. https://simonwillison.net/2025/Jul/29/space-invaders/

[ Parent | on HN ]
Re:Nvidia DGX Spark: great hardware, early days for the ecos... by magicalhippo on Wednesday October 15, 2025 @03:25AM (#45589129)
Depending on you use-case, I've been quite impressed with GPT-OSS 20B with high reasoning effort.

The 120B model is better but too slow since I only have 16GB VRAM. That model runs decent[1] on the Spark.

[1]: https://news.ycombinator.com/item?id=45576737

[ Parent | on HN ]
Re:Nvidia DGX Spark: great hardware, early days for the ecos... by cocogoatmain on Wednesday October 15, 2025 @03:30AM (#45589171)
128gb unified memory is enough for pretty good models, but honestly for the price of this it is better just go go with a few 3090s or a Mac due to memory bandwidth limitations of this card
[ Parent | on HN ]
Re:Nvidia DGX Spark: great hardware, early days for the ecos... by behnamoh on Wednesday October 15, 2025 @02:46AM (#45588875)
the question is: how does the prompt processing time on this compare to M3 Ultra because that one sucks at RAG even though it can technically handle huge models and long contexts...
[ Parent | on HN ]
Re:Nvidia DGX Spark: great hardware, early days for the ecos... by zozbot234 on Wednesday October 15, 2025 @04:43AM (#45589686)
Prompt processing time on Apple Silicon might benefit from making use of the NPU/Apple Neural Engine. (Note, the NPU is bad if you're limited by memory bandwidth, but prompt processing is compute limited.) Just needs someone to do the work.
[ Parent | on HN ]