NVIDIADesktop GPUCurrent

NVIDIA DGX Spark for local LLMs

Name: LLM Configurator — GPU VRAM Checker
Author: LLM Configurator

Written by Jakub Rusinowski · Last updated July 21, 2026

With 128 GB at 273 GB/s, the DGX Spark runs 141 catalogued models at Q4_K_M with 8K context. The largest that fits is Qwen3.8-Flash-Next (~109.7 GB), and the top pick is Qwen3.8-Flash-Next at 35–72 tok/s.

Compact AI development desktop built around the GB10 Grace Blackwell Superchip — a 20-core Grace ARM CPU plus a Blackwell GPU (6,144 CUDA cores) sharing 128 GB of unified LPDDR5x memory at 273 GB/s. Runs the full CUDA stack, unlike Apple Silicon, making it ideal for fine-tuning and inference of 70B-class models. Launched at $3,999 in October 2025; price raised to $4,699 in February 2026 due to memory supply constraints.

Check my hardware →Full benchmark table

Models that run on the DGX Spark

Q4_K_M, 8K context, 128 GB usable. Ranked by quality and speed.

Model	VRAM	Speed	Fit
Qwen3.8-Flash-Next Qwen3.8	109.7 GB	35–72 tok/s	Fits
Qwen 3.5 122B-A10B Qwen 3.5	74.5 GB	16–33 tok/s	Fits
Nemotron 3 Super 120B-A12B Nemotron 3 Super	73.3 GB	14–30 tok/s	Fits
Nemotron 3.5 Super VL 120B-A12B Nemotron 3.5	73.3 GB	14–30 tok/s	Fits
GPT-OSS 120B GPT-OSS	71.6 GB	38–80 tok/s	Fits
Llama 4 Scout 17B Llama 4	68.2 GB	12–25 tok/s	Fits
Llama 4.5 Scout Llama 4.5	66.6 GB	11–23 tok/s	Fits
Qwen3-Coder-Next (80B-A3B MoE) Qwen3-Coder	49.4 GB	58–120 tok/s	Fits
Kolibri 1 Kolibri	48.5 GB	52–109 tok/s	Fits
Nemotron 3.5 Bolt 74B Nemotron 3.5	45.6 GB	25–52 tok/s	Fits
K2 Horizon MoVA 36B-A4B K2 Horizon	25 GB	33–69 tok/s	Fits
Qwen 3.6 35B-A3B Qwen 3.6	22.2 GB	59–122 tok/s	Fits

Showing 12 of 141

Buy it or rent the same memory

Buy the card, or rent a GPU with the same memory by the hour to try models first.

Buy this hardware NVIDIA DGX Spark (128GB) — 128 GB VRAM · 150 W board power Deploy in the cloud now RunPod

or compare on Vast.ai

As an Amazon Associate we earn from qualifying purchases. Cloud GPU links are referral links — we may earn a commission at no extra cost to you.

Affiliate disclosure: Some links on this page are affiliate links — if you buy through them, LLM Configurator may earn a commission at no extra cost to you. As an Amazon Associate, LLM Configurator earns from qualifying purchases.

NVIDIA DGX Spark (128GB)

128 GB VRAM · 150 W board power

2026 prices are volatile — check the current listing.

Check price on Amazon

Speed vs other GPUs

Llama 3.1 8B, Q4_K_M. Estimated ranges. How this is calculated

H100 80GB (SXM)157–327 tok/s H100 80GB (PCIe)120–250 tok/s RTX 5090113–234 tok/s RTX PRO 6000 Blackwell113–234 tok/s RTX PRO 6000 Blackwell (Max-Q Edition)113–234 tok/s A100 80GB (PCIe)102–196 tok/s RTX PRO 5000 Blackwell (48 GB)94–194 tok/s RTX PRO 5000 Blackwell (72 GB)94–194 tok/s DGX Spark26–53 tok/s

Specifications

Specs last updated 2026-07-21.

Memory: 128 GB
Memory bandwidth: 273 GB/s
Architecture: GB10 Grace Blackwell Superchip
Series: AI Workstation
Board power: 150 W
Release year: 2025
Launch price: $4,699
Usable for models: 128 GB

Best for70B modelslocal LLM developmentfine-tuningunified memory

Recommended first model (Ollama)

Frequently asked questions

Can the NVIDIA DGX Spark run local LLMs?

Yes. With 128 GB (128 GB usable by a model) it runs 141 of the catalogued models at Q4_K_M with 8K context; the largest is Qwen3.8-Flash-Next, needing about 109.7 GB.

How fast is the NVIDIA DGX Spark for AI inference?

It is estimated to run Llama 3.1 8B at 26–53 tok/s at Q4_K_M. Llama 3.3 70B is estimated at 3.4–7.1 tok/s. These are modelled estimates from memory bandwidth, not measurements; the methodology page shows the formula.

What LLMs can I run on 128 GB?

Among the best that fit: Qwen3.8-Flash-Next, Qwen 3.5 122B-A10B, Nemotron 3 Super 120B-A12B, Nemotron 3.5 Super VL 120B-A12B, GPT-OSS 120B. The quickest start is Ollama: ollama run llama3.3:70b.

Can I run it on the DGX Spark?

More for this card

Tools

Guides