H200 SXM server

HGX H200, 8 SXM GPUs

Short answer. The H200 SXM server is an eight-GPU HGX board meshed through NVSwitch, giving a 1,128 GB memory pool. It makes the eight accelerators work as one, for training and inference of large models, with tensor parallelism under vLLM. The NVIDIA AI Enterprise licence is subscribed per GPU.

Hardware sheet · QDNA sovereign AI platform · Published on 30 July 2026 · Updated on 13 October 2026

Indicative price	about €451k excl. VAT: 8-GPU HGX H200 server with 141 GB per GPU and two Xeon Gold 6538Y+ listed at £387,785 excl. VAT by CTO Servers on 2 September 2026 (single vendor; the reseller asks for a quote to validate this price), converted at the ECB rate of 25 September 2026, £0.860 per €1 (price guide)
Memory	141 GB HBM3e per GPU, 8 GPUs (1,128 GB)
Bandwidth	4.8 TB/s per GPU
NVLink	900 GB/s per GPU, full NVSwitch mesh
Compute	3,958 TFLOPS FP8 and 1,979 TFLOPS FP16 per GPU with sparsity, tensor parallelism across 8 GPUs
Power	up to 700 W per GPU
Form factor	8-GPU HGX board, 8U server

What the machine runs

A 1128 GB pool for inference and training at scale: fine-tuning, LoRA and continued pre-training across eight GPUs meshed via NVLink. Recommended runtime: vLLM with tensor parallelism.

Positioning

The HGX H200 board makes eight GPUs work as a single accelerator via NVSwitch, for training and inference of large models. NVIDIA AI Enterprise licensing is sold per GPU.

Vendors

HGX H200 systems from Dell, HPE, Lenovo, Supermicro, ASUS and Gigabyte.

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 massive RAG at full speed.

Support

Open stack, maintained by its communities. Optional NVIDIA AI Enterprise (NIM, NeMo, Triton) with a service-level commitment, per GPU.

Which models fit on H200 SXM server?

The machine offers 1,128 GB of memory. 6 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 5.3	744	FP8	GLM 5.3 on H200 SXM server
Kimi K2.7 Code	1,000	NVFP4	Kimi K2.7 Code on H200 SXM server
DeepSeek V4	1,600	NVFP4	DeepSeek V4 on H200 SXM server
Nemotron 3 Ultra	550	FP8	Nemotron 3 Ultra on H200 SXM server
MiniMax M3	428	FP16	MiniMax M3 on H200 SXM server
Qwen 3.8 27B	27	FP16	Qwen 3.8 27B on H200 SXM server

See the full sizing matrix.

Frequently asked questions

How much does a configured h200 sxm cost for an LLM?

Pricing depends on configuration and supplier. The page shows a dated public price; a scoping call helps compare supplier quotes.

Which LLM fits in a h200 sxm?

Depends on the quantization format and the input context. The table on the page indicates the recommended maximum model size.

Which runtimes support the h200 sxm?

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.