DGX Station
Workstation, GB300 Desktop superchip
| Indicative price | €93,000 to €110,500 excl. VAT depending on the vendor (Supermicro and MSI €93,000, Exxact €95,000, Gigabyte €99,500, HP €110,000, ASUS €110,500), excluding VAT and shipping, as listed by the reseller pi3g on 27 August 2026; the Exxact Valence configuration at €95,000 excl. VAT is the value used in every calculation on the site (price guide) |
|---|---|
| Memory | 748 GB coherent: 252 GB HBM3e (7.1 TB/s) and 496 GB LPDDR5X (396 GB/s); 784 GB had been announced in March 2025 |
| GPU | Blackwell Ultra GB300 Desktop and Grace 72-core |
| Compute | 20 PFLOPS FP4 with sparsity (15 dense) |
| Network | ConnectX-8 SuperNIC, up to 800 Gb/s |
| Models | up to 1,000 billion parameters according to NVIDIA |
| Form factor | Workstation |
What the machine runs
Large models on premises with comfortable fine-tuning headroom. Recommended runtime: vLLM with Unsloth.
Positioning
SMBs get a full rack in a single chassis to develop and serve without a server room. A complete station is listed from €93,000 to €110,500 excl. VAT depending on the vendor (pi3g listing of 27 August 2026).
Manufacturers
The DGX Station is assembled by ASUS, Dell, Exxact, Gigabyte, HP, MSI and Supermicro around the GB300 board supplied by NVIDIA (partners listed by NVIDIA as of 2 September 2026).
GPUDirect Storage
On platforms fitted with ConnectX cards, GPUDirect Storage technology establishes a direct path between NVMe or NVMe over Fabric storage and GPU memory, bypassing the CPU buffer. Compatible storage arrays, such as NetApp, VAST, DDN or WEKA, feed training and large-scale RAG at full speed.
Support
Open stack supported by QDNA and the communities. Optional NVIDIA AI Enterprise (NIM, NeMo, Triton) with a service-level agreement, licensed per GPU.
Which models fit on DGX Station?
The machine offers 748 GB of memory. 5 of the 7 open models in the catalogue load on it, in the format shown.
| Model | Billion parameters | Most precise format that fits | Sizing page |
|---|---|---|---|
| GLM 5.2 | 744 | NVFP4 | GLM 5.2 on DGX Station |
| Kimi K2.7 Code | 1,000 | NVFP4 | Kimi K2.7 Code on DGX Station |
| Nemotron 3 Ultra | 550 | FP8 | Nemotron 3 Ultra on DGX Station |
| MiniMax M3 | 428 | FP8 | MiniMax M3 on DGX Station |
| Qwen 3.8 27B | 27 | FP16 | Qwen 3.8 27B on DGX Station |
See the full sizing matrix.
Frequently asked questions
How much does a configured dgx station cost for an LLM?
Pricing depends on configuration and supplier. The page shows a dated public price; we provide a detailed quote after a scoping call.
Which LLM fits in a dgx station?
Depends on the quantization format and the input context. The table on the page indicates the recommended maximum model size.
Which runtimes support the dgx station?
vLLM, llama.cpp, Triton Inference Server, depending on the chosen framework. The model, format and GPU combinations measured by QDNA are published in the measurements section.