# Workstation PNY NVIDIA DGX Spark GB10 — 128 GB LPDDR5x, 4 TB SSD, NVIDIA Blackwell, DGX OS, Wi‑Fi, 240W

**URL:** https://aio.lv/en/product--pny-940-54242-0005-000--12389156
**Data updated:** 2026-04-26 16:17 (EEST)
**Category:** Desktop PCs & Workstations
**Manufacturer:** PNY
**Manufacturer code:** 940-54242-0005-000
**EAN:** 810152850282
**Warranty:** 2 Years
**Availability:** Available for order
**Price:** 5 689.49 €
**Prices and availability valid until:** 2026-08-15 23:59
**Price without VAT (21%):** 4 702.06 EUR
**Can be purchased by legal entities:** Yes (invoice)
**Buy:** https://aio.lv/en/product--pny-940-54242-0005-000--12389156

## Delivery options

| Delivery method | Delivery time | Cost |
|---|---|---|
| Office pickup (2b. Kengaraga str., Riga, Latvia, LV-1063) | 30.07.2026 | free |
| Pickup points | 04.08.2026 | 1.89 EUR |
| Courier delivery to door | 04.08.2026 | 4.99 EUR |

## Summary description

PNY NVIDIA DGX Spark GB10 workstation with GB10 processor, 128 GB LPDDR5x system memory and 4 TB SSD storage. Equipped with onboard NVIDIA Blackwell graphics, DGX OS preinstalled, Wi‑Fi connectivity and a 240 W power supply. Designed for AI development, data science workloads, model training and edge or compact deployment where high memory and fast NVMe SSD storage are required.

## Long description

The PNY NVIDIA DGX Spark GB10 is a compact AI workstation powered by the NVIDIA GB10 Grace Blackwell Superchip. It combines a GB10 processor with 128 GB of coherent LPDDR5x system memory, a 4 TB NVMe SSD, onboard NVIDIA Blackwell graphics and DGX OS preinstalled to support local development, fine‑tuning and inference of large reasoning models. The system targets developers and data scientists who need a power‑efficient, small‑form AI platform for prototyping and running models with up to hundreds of billions of parameters.
Advantages
The workstation delivers up to one petaFLOP of AI performance in a compact, desk‑side form factor and includes a full NVIDIA AI software stack for immediate use. High‑capacity LPDDR5x unified memory and fast NVMe storage enable large in‑memory models and rapid dataset access, while integrated networking and a 240 W power supply support sustained workloads.
Suitable for
Local development and prototyping of large language and reasoning models
Fine‑tuning and inference of models up to ~200 billion parameters
Data science workloads and experiment workflows requiring large system memory
Edge or compact deployments where space and power are limited
When it may not be suitable
This compact DGX Spark is not intended to replace large multi‑node GPU clusters for massive distributed training at datacenter scale. For workloads that require multiple discrete high‑end GPUs or extensive horizontal scaling, a rack or datacenter solution will be more appropriate. It is optimized for memory‑heavy, single‑node workflows rather than large multi‑server training runs.
Key specifications
Processor: NVIDIA GB10 Grace Blackwell Superchip with 20 GB10 cores
System memory: 128 GB LPDDR5x coherent unified memory
Storage: 4 TB NVMe SSD
Graphics: NVIDIA Blackwell integrated
Networking: Ethernet up to 10 Gbit/s, Wi‑Fi 7 (802.11be), Bluetooth 5.4
Power supply: 240 W
Software: DGX OS with NVIDIA AI software stack preinstalled
How to use
Set up the DGX Spark by connecting power and network, then follow the DGX OS initial configuration prompts. Use the preinstalled NVIDIA AI software stack and containerized tooling to deploy, train and run inference on models. Allocate large models to the unified LPDDR5x memory and use the NVMe SSD for datasets and checkpoints; monitor thermals and power usage during sustained workloads to maintain performance.
What to consider before purchase
Confirm that a single‑node workstation with high unified memory meets your model size and workload requirements
Evaluate whether 240 W power and the compact form factor fit your deployment environment
Check networking needs — integrated 10 Gbit/s Ethernet and Wi‑Fi 7 may or may not match your infrastructure
Plan storage and backup strategy for the 4 TB NVMe SSD
Usage recommendations
For development and inference of large reasoning models, keep datasets and checkpoints on the NVMe SSD and run active model parameters in the LPDDR5x unified memory. Use the preinstalled NVIDIA AI stack and container workflows for reproducible experiments, and monitor system temperature and power during extended runs to avoid thermal throttling.

## Specifications

### Processor

| Feature | Value |
|---|---|
| Processor model | GB10 |
| Processor cores | 20 |
| Processor manufacturer | Nvidia |

### Memory

| Feature | Value |
|---|---|
| Internal memory | 128 GB |
| Internal memory type | LPDDR5x-SDRAM |

### Storage

| Feature | Value |
|---|---|
| Total storage capacity | 4000 GB |
| Storage media | SSD |
| SSD form factor | M.2 |
| NVMe | Yes |

### Graphics

| Feature | Value |
|---|---|
| Discrete graphics card model | Not available |
| Discrete graphics card | No |
| On-board graphics card | Yes |
| On-board GPU manufacturer | NVIDIA |

### Network

| Feature | Value |
|---|---|
| Wi-Fi | Yes |
| Bluetooth | Yes |
| Ethernet LAN | Yes |
| Bluetooth version | 5.4 |
| Ethernet LAN data rates | 10000 |
| Top Wi-Fi standard | Wi-Fi 7 (802.11be) |
| Bluetooth Low Energy (BLE) | Yes |

### Ports & interfaces

| Feature | Value |
|---|---|
| HDMI ports quantity | 1 |
| HDMI version | 2.1a |
| USB 3.2 Gen 1 (3.1 Gen 1) Type-C ports quantity | 4 |

### Software

| Feature | Value |
|---|---|
| Operating system installed | DGX OS |

### Design

| Feature | Value |
|---|---|
| Product colour | Gold |

### Weight & dimensions

| Feature | Value |
|---|---|
| Weight | 1.2 kg |
| Height | 50.5 mm |
| Width | 150 mm |
| Depth | 150 mm |

### Other features

| Feature | Value |
|---|---|
| Power supply | 240 W |
| Display included | No |

### Performance

| Feature | Value |
|---|---|
| Optical drive type | No |
| Product type | Workstation |

### Neural processor unit (NPU)

| Feature | Value |
|---|---|
| Copilot+ PC | No |

## FAQ

### How much does the PNY NVIDIA DGX Spark GB10 workstation cost, and is it available for purchase?

The price is 5,689.49 € (valid at the time of publication and already includes all taxes). The item is available for order — 6 pcs. are in stock, and you can place your purchase right now with delivery across Latvia and Europe or choose pickup at the Riga collection point. We offer favorable purchase terms, including the option to split the payment into convenient monthly installments. Availability and price are updated in real time, so we recommend placing your order in advance.

### If I order this product today, when will I receive my order?

If you place your order today, delivery will be as fast and convenient as possible. You will be able to pick up your order on 30.07.2026 at the Riga pickup point or choose delivery across Latvia and Europe. We carefully inspect every purchase and pack it securely for safe transportation.

### What payment options are available?

AiO.lv offers several payment methods: Bank transfer, Leasing for legal entities. All transactions are 100% secure and encrypted.

### Can this product be purchased for a company (with VAT)?

Yes. AiO.lv works with legal entities in Latvia and across Europe. The product price already includes VAT. For certain product categories, as well as for companies from other EU countries with a valid VAT number, a 0% VAT rate may apply. We issue the invoice immediately after the order is placed, support electronic BIS invoices, and automatically generate all necessary accounting documents. You can complete the purchase for a company online — simply select the company profile and choose bank transfer as the payment method during checkout.

### What is the warranty period?

All PNY products come with an official 2 Years warranty. It covers all manufacturing defects and malfunctions that may occur during use. The warranty is valid at all authorized PNY service centers and through AiO.lv

### Can the product be returned after purchase?

Yes, you can return the product within 14 days of purchase if you bought it as a consumer (not a legal entity). Returns are possible without giving any reason — the main condition is that the product retains its original appearance and full set of accessories. We will make the process as simple and convenient as possible: just contact our customer support team, and we will help you arrange the return.

### Where should I contact for warranty service?

For warranty-related questions you can contact the AiO.lv (SIA VSPRO) customer support team. More information about warranty service: https://aio.lv/en/warranty

### How can I tell if this product suits my needs?

When making a choice, we recommend paying attention to the key features: processor model, internal memory, internal memory type, discrete graphics card model, operating system installed, weight, product colour, compatibility with your devices or usage conditions, material quality, functionality, and warranty terms. These characteristics will help you determine how well the model meets your needs.

### Where can I download the instructions for PNY NVIDIA DGX Spark GB10 workstation?

You can download the instructions in PDF format from the “Instructions” section on the product description page. The file contains complete information about installation, setup, and usage.

## Image

https://aio.lv/img/cache/product/12389156/68552054_large.webp