text model · Nemotron · Linux

Can I run Nemotron 3.5 Super VL 120B-A12B on Nvidia DGX Spark (128GB)?

Compatibility verdict VRAM threshold engine

Yes, it runs GPU accelerated

Yes. Nemotron 3.5 Super VL 120B-A12B runs on Nvidia DGX Spark (128GB) at Q4_K_M (~87.4 GB of ~112 GB usable).

Needs ~87.4 GB Device usable ~112 GB

Runs at Q4_K_M using ~87.4 GB of ~112 GB usable.

That figure is at a 4k context and moves about ±15% as context length changes. Nvidia DGX Spark (128GB) leaves ~24.6 GB of headroom.

Q4_K_M needed: ~87.4 GB
Usable on device: ~112 GB
Device memory: 128 GB
Best quant: Q4_K_M

Share on X Share on Reddit

Which quant fits

Quant ladder vs ~112 GB usable

Q2_K

~53.7 GB

Q3_K_M

~62.1 GB

Q4_K_M

~87.4 GB

Q5_K_M

~88.9 GB

Q6_K

~101.8 GB

Q8_0

~130.9 GB

FP16

~243.4 GB

The line marks Nvidia DGX Spark (128GB)'s ~112 GB budget; rungs past it are too large.

Run it

Install commands Linux

Loads the Q4_K_M weights.

Ollama

$ ollama run nemotron-3.5-super:120b-a12b-q4_K_M

Model Nemotron

Parameters: 120B (MoE, 12B active)
Q4_K_M size: 83.96 GB
Context: 1024k
Ollama tag: nemotron-3.5-super:120b-a12b-q4_K_M

Full Nemotron 3.5 Super VL 120B-A12B requirements →

Device Linux

Memory: 128 GB unified
Usable for weights: ~112 GB
Power draw: ~140 W
Best runtime: llama.cpp (CUDA) / Ollama (CUDA)

Best models for Nvidia DGX Spark (128GB) →

You could also run

Run Nemotron 3.5 Super VL 120B-A12B on other hardware

Apple M4 Max (128GB)AMD Ryzen AI Halo (128GB)Apple M5 Max (128GB)Apple M3 Ultra (256GB)

FAQ

Can Nvidia DGX Spark (128GB) run Nemotron 3.5 Super VL 120B-A12B?

Yes. Nemotron 3.5 Super VL 120B-A12B runs on Nvidia DGX Spark (128GB) at Q4_K_M (~87.4 GB of ~112 GB usable).

How much memory does Nemotron 3.5 Super VL 120B-A12B need?

Nvidia DGX Spark (128GB) has room to spare. At Q4_K_M the weights are ~83.96 GB; with KV cache and runtime overhead, budget ~87.4 GB at a 4k context. It is a Mixture-of-Experts model (120B total / 12B active), so all experts must stay in memory; memory tracks total params, not active params.

What is the best tool to run Nemotron 3.5 Super VL 120B-A12B on Linux?

Ollama for a simple setup; vLLM for the most speed. vLLM shines for multi-user serving/throughput. For a single local chat, Ollama or llama.cpp is simpler and lighter.

Embed this

A live badge for your model card or README, updated as the data is.

Markdown

[![Nemotron 3.5 Super VL 120B-A12B on Nvidia DGX Spark (128GB)](https://localmodel.run/badge/nemotron-3.5-super-120b-a12b/nvidia-dgx-spark-128gb.svg)](https://localmodel.run/can-i-run/nemotron-3.5-super-120b-a12b/nvidia-dgx-spark-128gb)

Sources

docs.nvidia.com · 2 sources
- docs.nvidia.com/dgx/dgx-spark
- docs.nvidia.com/dgx/dgx-spark-porting-guide
forums.developer.nvidia.com · 2 sources
- forums.developer.nvidia.com/t/difference-in-total-vram-available-for-different-sparks
- forums.developer.nvidia.com/t/memory-creep-on-dgx-spark-where-your-128-gb-actually-goes-and-how-to-stop-it
github.com · 1 source
- github.com
huggingface.co · 2 sources
- huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16
- huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16/blob
nvidia.com · 1 source
- nvidia.com
nvidianews.nvidia.com · 1 source
- nvidianews.nvidia.com
ollama.com · 2 sources
- ollama.com
- ollama.com/library
registry.ollama.ai · 1 source
- registry.ollama.ai
tomshardware.com · 1 source
- tomshardware.com

Weights are measured from GGUF files; KV cache and overhead are computed, so totals can vary ~15% with context and runtime. Any tok/s is a bandwidth estimate. See methodology.