# Anyone Using the NVIDIA DGX Spark for Comfy?

**URL:** <https://forum.logik.tv/t/anyone-using-the-nvidia-dgx-spark-for-comfy/13806>\
**Category:** Hardware Questions\
**Tags:** jobs\
**Created:** [November 19, 2025, 3:22pm UTC](https://forum.logik.tv/t/anyone-using-the-nvidia-dgx-spark-for-comfy/13806 "2025-11-19T15:22:25Z")\
**Posts on this page:** 10\
**Page:** 1

<div class="post-metadata">

**Author:** ![RufusBlackwell](https://sea2.discourse-cdn.com/flex020/user_avatar/forum.logik.tv/rufusblackwell/32/422_2.png) [@RufusBlackwell](https://forum.logik.tv/u/RufusBlackwell)\
**Post date:** [November 19, 2025, 3:22pm UTC](https://forum.logik.tv/t/anyone-using-the-nvidia-dgx-spark-for-comfy/13806/1 "2025-11-19T15:22:25Z")

</div>

I’m running Flame on a Mac Studio and Comfy in the cloud. Good in some ways, annoying in others. Thinking about getting the Nvidia Spark as a standalone Comfy workstation next to my Flame. Anyone doing anything like this, would be very interested in any insights.

Cool little machine. Not the fastest I think, but the massive 128gb unified memory would be v useful large video models. And can double it up for 256gb later maybe

[https://www.nvidia.com/en-us/products/workstations/dgx-spark/](https://www.nvidia.com/en-us/products/workstations/dgx-spark/)

[https://www.amazon.com/gp/product/B0FWJ16CCH/ref=ox\_sc\_act\_title\_1?smid=ASF0S1GFIHF5V&psc=1](https://www.amazon.com/gp/product/B0FWJ16CCH/ref=ox_sc_act_title_1?smid=ASF0S1GFIHF5V&psc=1)

---

<div class="post-metadata">

**Author:** ![julioleon](https://sea2.discourse-cdn.com/flex020/user_avatar/forum.logik.tv/julioleon/32/14482_2.png) [@julioleon](https://forum.logik.tv/u/julioleon)\
**Post date:** [November 19, 2025, 4:23pm UTC](https://forum.logik.tv/t/anyone-using-the-nvidia-dgx-spark-for-comfy/13806/2 "2025-11-19T16:23:55Z")

</div>

Does it runs Windows?

---

<div class="post-metadata">

**Author:** ![RufusBlackwell](https://sea2.discourse-cdn.com/flex020/user_avatar/forum.logik.tv/rufusblackwell/32/422_2.png) [@RufusBlackwell](https://forum.logik.tv/u/RufusBlackwell)\
**Post date:** [November 19, 2025, 4:29pm UTC](https://forum.logik.tv/t/anyone-using-the-nvidia-dgx-spark-for-comfy/13806/3 "2025-11-19T16:29:55Z")

</div>

As far as I understand it, it runs on it’s own OS, but its basically Linux

---

<div class="post-metadata">

**Author:** ![randy](https://sea2.discourse-cdn.com/flex020/user_avatar/forum.logik.tv/randy/32/6122_2.png) [@randy](https://forum.logik.tv/u/randy)\
**Post date:** [November 20, 2025, 2:46am UTC](https://forum.logik.tv/t/anyone-using-the-nvidia-dgx-spark-for-comfy/13806/4 "2025-11-20T02:46:34Z")

</div>

It feels like it’s impressive as a V ram but not in a compute perspective. Most of the influencers on YouTube I’ve reviewed it. Check out network Chuck’s on YouTube. He had a review about a month back. After watching that I was like yeah I think we still need to stick with big ass beefy Nvidia 5000 Adas or 6000 pro Blackwells.

---

<div class="post-metadata">

**Author:** ![RufusBlackwell](https://sea2.discourse-cdn.com/flex020/user_avatar/forum.logik.tv/rufusblackwell/32/422_2.png) [@RufusBlackwell](https://forum.logik.tv/u/RufusBlackwell)\
**Post date:** [November 20, 2025, 3:06am UTC](https://forum.logik.tv/t/anyone-using-the-nvidia-dgx-spark-for-comfy/13806/5 "2025-11-20T03:06:48Z")

</div>

> [@randy](#):
>
> or 6000 pro Blackwells.

Jesus Christ, not sure anyone needs 6000 of me!

---

<div class="post-metadata">

**Author:** ![randy](https://sea2.discourse-cdn.com/flex020/user_avatar/forum.logik.tv/randy/32/6122_2.png) [@randy](https://forum.logik.tv/u/randy)\
**Post date:** [November 20, 2025, 4:05am UTC](https://forum.logik.tv/t/anyone-using-the-nvidia-dgx-spark-for-comfy/13806/6 "2025-11-20T04:05:04Z")

</div>

![Argue Seth Meyers GIF by Late Night with Seth Meyers](https://us1.discourse-cdn.com/flex020/uploads/thedepartmentofexternalservices/original/2X/8/8dd57f7c39aa068300cbbf0a4152cfbc2082dcc7.gif)

---

<div class="post-metadata">

**Author:** ![allklier](https://sea2.discourse-cdn.com/flex020/user_avatar/forum.logik.tv/allklier/32/4309_2.png) [@allklier](https://forum.logik.tv/u/allklier)\
**Post date:** [November 20, 2025, 6:45am UTC](https://forum.logik.tv/t/anyone-using-the-nvidia-dgx-spark-for-comfy/13806/7 "2025-11-20T06:45:38Z")

</div>

I’ll second that. Was running some ComfyUI models locally on A6000 Classic (pre Ada), and then on RunComfy H200. Problem was not Vram constrained, but compute constrained. H200 was only 4x faster. Now I’ll take that any day, but the use cases may vary, mind the details.

Bigger issue is that some of these tools don’t scale horizontally, which is where it seems the big custom cloud models have an edge, because they are custom engineered between software and hardware.

---

<div class="post-metadata">

**Author:** ![Sinan](https://sea2.discourse-cdn.com/flex020/user_avatar/forum.logik.tv/sinan/32/155_2.png) [@Sinan](https://forum.logik.tv/u/Sinan)\
**Post date:** [November 20, 2025, 7:39am UTC](https://forum.logik.tv/t/anyone-using-the-nvidia-dgx-spark-for-comfy/13806/8 "2025-11-20T07:39:01Z")

</div>

That would probably cover the total Flame artist requirement in the entire world.

---

<div class="post-metadata">

**Author:** ![kily](https://avatars.discourse-cdn.com/v4/letter/k/a88e4f/32.png) [@kily](https://forum.logik.tv/u/kily)\
**Post date:** [November 20, 2025, 10:49am UTC](https://forum.logik.tv/t/anyone-using-the-nvidia-dgx-spark-for-comfy/13806/9 "2025-11-20T10:49:45Z")

</div>

Mediocre performance, interesting only because of the unified memory concept, run its own OS… could we slap an apple sticker on it, 4 wheels by 700$, remove ports, and no one would notice?

---

<div class="post-metadata">

**Author:** ![marios](https://avatars.discourse-cdn.com/v4/letter/m/f04885/32.png) [@marios](https://forum.logik.tv/u/marios)\
**Post date:** [November 20, 2025, 3:38pm UTC](https://forum.logik.tv/t/anyone-using-the-nvidia-dgx-spark-for-comfy/13806/10 "2025-11-20T15:38:05Z")

</div>

I saw this the other day seems like a good review to help decide …. [https://www.youtube.com/watch?v=FYL9e\_aqZY0](https://www.youtube.com/watch?v=FYL9e_aqZY0)
