Crashes, No Sound, or Screen Glitches?
Random freezes, missing sound and display glitches usually trace back to one bad driver. Find and replace yours safely.Free scan · under a minutePC Slower Than It Used to Be?
A free scan shows the junk files, broken settings and background clutter dragging Windows down - then fixes them in one click.Free scan · Windows 10 & 11iTechGuides is reader-supported. When you buy through links on our site, we may earn an affiliate commission. As an Amazon Associate I earn from qualifying purchases. Learn more
NVIDIA has announced a Windows version of its GB300-based DGX Station, a deskside AI system that pairs Windows support with access to Linux AI toolchains through Windows Subsystem for Linux (WSL). The headline figures are up to 748 GB of coherent CPU-and-GPU memory and up to 20 PFLOPS of FP4 Tensor Core compute with sparsity. Neither figure means 748 GB of GPU VRAM or a generic performance guarantee.
The Windows model is not yet listed as shipping: NVIDIA’s product page says “Coming in Q4,” and its June 1, 2026 announcement says systems from six OEMs are expected in Q4 2026. Exact configurations, prices and delivery dates have not been established in the cited announcements.
| # | Preview | Product | Price | |
|---|---|---|---|---|
| 1 |
|
UGREEN Maxidok Thunderbolt 5 Docking Station, 8TB M.2 NVMe SSD Support | $499.99 | Buy on Amazon |
| 2 |
|
NVIDIA RTX 4000 SFF Ada Generation Workstation Ada Lovelace Architecture Dual Slot Low Profile... | $1,990.00 | Buy on Amazon |
| 3 |
|
Nvidia RTX 4000 Ada Retail | $1,995.00 | Buy on Amazon |
What changed: DGX Station now has a Windows path
NVIDIA’s announcement adds a Windows-oriented software configuration to its GB300-based DGX Station platform. NVIDIA describes WSL as the way to access Linux AI toolchains on the Windows system. That is not the same as saying Linux has been eliminated, that all Linux applications run natively in Windows, or that every Linux workload will work in WSL.
NVIDIA’s DGX Station Development Guide documents an Ubuntu 24.04 software environment. The available material describes distinct Windows and Ubuntu paths; it does not establish that a single machine supports dual boot. Buyers should confirm the operating system and supported software stack for the exact OEM configuration.
#1 Best Overall
- All-in-One Connectivity: Build a tidy workstation with 17 ports: 3×USB-A 10Gbps, 3×USB-C 10Gbps (2 ports share 60W charging), SD/TF 4.0, 3×3.5mm audio, 2.5G Ethernet, DP1.4, and 2×Thunderbolt 5 downstream ports
- 120Gbps Thunderbolt 5 Speed & M.2 High-Speed Storage: Intel-certified Thunderbolt 5 provides up to 120Gbps bandwidth. Built-in NVMe PCIe Gen4x4 M.2 slot supports 2230–2280 SSDs up to 8TB for fast external storage. Note: Compatible with M.2 NVMe SSDs with heatsink (height≤12 mm; width≤24 mm), Not compatible with SSDs wider than 24 mm, including Seagate FireCuda 530 with Heatsink, Samsung 9100 PRO with Heatsink, WD_BLACK SN850X Premium Model with OEM Heatsink, and Lexar NM1090 Active Fan Version
- Multi-Monitor 8K/6K Output: Supports triple independent displays on Windows (2×TBT5+DP): dual or single 8K@60Hz. Mac supports dual displays. Native compatible with latest Windows and MacBooks for expanded workspace. Note: The DP port does not support DP++. To connect an HDMI monitor, please use an active DP-to-HDMI adapter or cable; passive adapters and cables will not work
- 240W Total System Power & High-Speed Networking (required) Equipped with up to 240W total power and a built-in 2.5GbE network port for fast and reliable wired connections, with lower latency and a more stable connection
- AI Cooling & Aluminum Chassis: AI smart cooling with 60mm ultra-thin fan ensures stable performance under heavy load. Premium aluminum unibody offers great heat dissipation, scratch resistance, and a clean matte finish, paired with a reliable security lock. Note: It is normal for the cooling fan to make noise when it is running and the noise is below 30 dB (data from UGREEN lab)
What 748 GB of coherent memory means
NVIDIA’s advertised total of up to 748 GB combines two types of memory: 252 GB of HBM3e GPU memory and 496 GB of LPDDR5X CPU memory. It is a coherent CPU-and-GPU memory pool, not 748 GB of GPU VRAM. The distinction matters when estimating whether a model or workload can fit and how it will use the system’s memory hierarchy.
The platform centers on the GB300 Grace Blackwell Ultra Desktop Superchip: a 72-core Grace CPU connected to a Blackwell Ultra GPU by NVLink-C2C. NVIDIA lists that interconnect at 900 GB/s. These are platform specifications, not a guarantee that every application will make equally effective use of the available memory or bandwidth.
How to read the 20 PFLOPS figure
The headline maximum is up to 20 PFLOPS of FP4 Tensor Core compute with sparsity, according to NVIDIA’s 2026 product information. NVIDIA lists 15 PFLOPS for FP4 without sparsity. The sparsity and precision qualifiers are essential: 20 PFLOPS is not an all-purpose measure of workstation speed, nor a prediction of performance for every model or task.
Rank #2
- VD8465 Japanese Authorized Distributor Product
- The speed of FP32 calculation is twice as fast as previous generations, which greatly improves the complex 3D processing and graphics simulation workflow
- Up to 2X the throughput compared to previous generations and significantly faster workloads such as video content rendering, architectural design assessments, and virtual prototypes of product design
- Achieve more than twice the previous generation AI performance improvement, support faster FP8 precision data and accelerate the execution of mixed flotation decimal and whole numbers
- It has a large capacity of memory necessary for working with a vast array of data sets and workloads such as rendering, data science, and simulation
| NVIDIA-listed compute | Qualification |
|---|---|
| Up to 20 PFLOPS | FP4 Tensor Core compute with sparsity |
| 15 PFLOPS | FP4 Tensor Core compute without sparsity |
| 10 PFLOPS | FP8/FP6 |
| 5 PFLOPS | FP16/BF16 |
| 80 TFLOPS | FP32 |
NVIDIA says the system supports models of up to one trillion parameters and targets local AI development, fine-tuning, inference, data science and AI agents connected to Windows applications. Treat the model-capacity statement and compute figures as vendor claims; the cited official material does not provide an independent benchmark of the Windows variant or establish performance for an arbitrary workload.
Power, networking and optional graphics to check
NVIDIA lists total system power at 1,600 W and a 20 A circuit requirement. Those figures make electrical capacity a practical deployment check, especially for a deskside installation. The system also lists ConnectX-8 SuperNIC networking of up to 800 Gb/s; that maximum does not guarantee a particular transfer rate in a real network.
NVIDIA lists several supported optional RTX PRO cards: RTX PRO 6000 Workstation Edition, RTX PRO 6000 Blackwell Max-Q Workstation Edition, RTX PRO 4000 Blackwell SFF Edition and RTX PRO 2000 Blackwell. NVIDIA says an RTX PRO 6000 pairing can add ray-traced visualization and simulation. The RTX PRO option is separate from the GB300 superchip, and NVIDIA notes that support varies by system. Confirm the exact supported card and SKU with the OEM rather than assuming every model is available in every workstation.
Rank #3
- NVIDIA Quadro Sync II1 compatibility
- 3D stereo support with stereo connector
- NVIDIA GPUDirect for Video support
- NVIDIA GPUDirect Remote Direct Memory Access (RDMA) support
- NVIDIA RTX ExperienceTM
When will the Windows DGX Station be available?
NVIDIA’s Windows product page currently says “Coming in Q4.” Its June 1, 2026 announcement names ASUS, Dell Technologies, GIGABYTE, HP, MSI and Supermicro as expected providers in Q4 2026. That is an expected availability window, not a confirmed shipping date or proof that all six will release systems at the same time.
Free tools Windows power users keep installed
One-click scans. No signup required.
Microsoft’s October 7, 2026 Windows Experience Blog separately says Dell Pro Precision with GB300 and HP ZGX Fury AI Station are coming later in 2026. The cited announcements do not establish exact order dates, prices or final Windows configurations. Check the relevant OEM’s listing for regional availability and confirmed specifications before making a purchase decision.
What to compare when OEM listings appear
Once manufacturers publish complete configurations, compare the specific system rather than assuming all DGX Station offerings are identical. Check:
- The exact GB300 configuration and included memory and storage.
- Whether an RTX PRO GPU is included or optional, and which exact card the SKU supports.
- The Windows edition, WSL support and validated AI software stack; compare with the Ubuntu configuration if Linux is required.
- Networking, including the installed adapter and the equipment needed to use its rated capacity.
- Power draw, the stated circuit requirement and whether the intended workspace can meet it.
- Warranty, enterprise support, regional availability, delivery timing and price.
These details determine whether a configuration fits a particular workflow and deployment. NVIDIA’s platform-level specifications alone do not settle those OEM-specific questions.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
Recommended Free Tools
