QDNASales and integration of LLM inference and training platforms, on-premises or hybrid

Running GLM 5.2 on H200 SXM server

GLM 5.2 has 744 billion parameters. H200 SXM server offers 1,128 GB of HBM3e across 8 GPUs memory. This page puts the two side by side.

Short answer. GLM 5.2 fits comfortably on H200 SXM server. In FP8, its weights take about 856 GB of the 1,128 GB available, leaving 272 GB for the attention cache and concurrency.

How much memory does GLM 5.2 need?

Weights take the parameter count multiplied by the format size, plus a 15% runtime margin for activations and buffers. The attention cache sits on top: it grows with context length and with the number of concurrent requests, so it is sized case by case.

FormatWeights in memory On this machineWhat the format costs
FP161,711 GBdoes not fitfull precision, the quality reference
FP8856 GBtightnegligible loss on most tasks
NVFP4428 GBfitsBlackwell format, native FP4 compute
Q4428 GBfitsblock quantisation, broadly supported

What is left to serve requests?

GLM 5.2 fits comfortably on H200 SXM server. In FP8, its weights take about 856 GB of the 1,128 GB available, leaving 272 GB for the attention cache and concurrency. The remaining margin decides how many concurrent requests and how much context the machine sustains. A margin below a quarter of memory forces a limit on context or on concurrency.

H200 SXM server ranges from 350 to 500 k€, indicative and not contractual. The machine targets the GE segment.

Where else can GLM 5.2 run?

See the GLM 5.2 and H200 SXM server fact sheets.

How much memory does GLM 5.2 need?

In FP8, weights take about 856 GB including the runtime margin. The attention cache sits on top and depends on context.

Does GLM 5.2 fit on H200 SXM server?

GLM 5.2 fits comfortably on H200 SXM server. In FP8, its weights take about 856 GB of the 1,128 GB available, leaving 272 GB for the attention cache and concurrency.

Method

Parameter counts and memory figures come from the site fact sheets. Memory footprints are calculated, not measured: a reading on real hardware may differ depending on the engine and the exact weight format. Prices are indicative and not contractual.