Running GLM 5.2 on B200 SXM

GLM 5.2 has 744 billion parameters. B200 SXM offers 1,440 GB of HBM3e across 8 GPUs memory. This page puts the two side by side.

By QDNA · Published on 27 July 2026 · Updated on 15 September 2026

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. In FP8 the calculation reads: 744 billion parameters × 1 byte (FP8) = 744.0 GB of weights; × 1.15 runtime margin = 855.6 GB. 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.

Model sheet: 744 billion parameters, 40 billion active per token, 1,048,576-token context, MIT licence. 744 billion announced by Z.AI; the repository holds 753 billion tensor elements in BF16 (1,507 GB).

GLM 5.2 activates 40 billion parameters per token out of 744 billion. Active parameters govern speed, not footprint: every expert stays resident in memory. Confusing the two under-sizes the machine by an order of magnitude.

What is left to serve requests?

GLM 5.2 fits comfortably on B200 SXM. In FP8, its weights take about 856 GB of the 1,440 GB available, leaving 584 GB for the attention cache and concurrency. The remaining margin decides how many concurrent requests and how much context the machine sustains. By QDNA convention a machine “fits” when the weights take at most 75% of its memory; beyond that it “barely fits” and forces a limit on context or on concurrency. The attention cache is not calculated on this page. These models’ architectures (compressed latent, linear or sparse attention, Mamba layers) share no per-token formula; it is measured on the machine, with the target context and concurrency [TO BE MEASURED].

B200 SXM: price on quotation. NVIDIA publishes no price; Aivres quoted about $340,000 excl. VAT for an HGX B200 server (Arc Compute, updated 30 July 2026, flagged as outdated by the author). The machine targets the GE segment.

Where else can GLM 5.2 run?

• GLM 5.2 on Mac Studio Ultra
• GLM 5.2 on DGX Station
• GLM 5.2 on RTX PRO 6000 server
• GLM 5.2 on H200 SXM server
• GLM 5.2 on B300 SXM

See the GLM 5.2 and B200 SXM fact sheets.

Method and sources

Memory footprints are calculated, not measured: a reading on real hardware may differ depending on the engine and the exact weight format. Every figure below carries its source and the date of its last check.

• GLM 5.2: 744 billion parameters, 40 billion active, 1,048,576-token context, MIT licence. Hugging Face repository (config.json, API, model card), checked on 2 September 2026. 744 billion announced by Z.AI; the repository holds 753 billion tensor elements in BF16 (1,507 GB).
• B200 SXM: 1,440 GB of memory; HGX B200, 8 Blackwell GPUs, 1.4 TB total memory, 144 PFLOPS FP4 with sparsity (72 dense), 1.8 TB/s NVLink per GPU. Manufacturer sheet checked on 2 September 2026.
• B200 SXM price: on quotation. NVIDIA publishes no price; Aivres quoted about $340,000 excl. VAT for an HGX B200 server (Arc Compute, updated 30 July 2026, flagged as outdated by the author).
• Bytes per parameter: FP16 2 (16 bits per parameter, hence 2 bytes); FP8 1 (8 bits per parameter (E4M3), hence 1 byte, block scales not counted); NVFP4 0.5 (4 bits per parameter, hence 0.5 byte; 4-bit published weights weigh 0.54 to 0.56 byte per parameter with their scales (DeepSeek V4 Pro 865 GB for 1,599 billion, Kimi K3 1,561 GB for 2,780 billion)); Q4 0.6 (taken here as llama.cpp Q4_K_M, 0.6 byte per parameter including scales; Q4_K_S weighs 0.56, AWQ and GPTQ 0.55). Source: connaissance/faits.yaml, quantifications family, checked on 2026-08-31.
• Runtime margin × 1.15: QDNA operating assumption, not measured: 15% above the weights for activations, buffers and fragmentation.
• “Fits” threshold at 75% of memory: QDNA assumption, not measured, which keeps the remaining quarter for the attention cache and concurrency.