QDNAAdvisory and architecture for LLM inference and training platforms, on-premises or hybrid

Running Qwen 3.8 27B on DGX Spark

Qwen 3.8 27B has 27 billion parameters. DGX Spark offers 128 GB of unified LPDDR5X memory. This page puts the two side by side.

By QDNA · Published on · Updated on

Short answer. Qwen 3.8 27B fits comfortably on DGX Spark. In FP16, its weights take about 62.1 GB of the 128 GB available, leaving 65.9 GB for the attention cache and concurrency.

How much memory does Qwen 3.8 27B need?

Weights take the parameter count multiplied by the format size, plus a 15% runtime margin for activations and buffers. In FP16 the calculation reads: 27 billion parameters × 2 bytes (FP16) = 54.0 GB of weights; × 1.15 runtime margin = 62.1 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.

FormatWeights in memory On this machineWhat the format costs
FP1662.1 GBfitsfull precision, the quality reference
FP831.0 GBfitsnegligible loss on most tasks
NVFP415.5 GBfitsBlackwell format, native FP4 compute
Q418.6 GBfitsblock quantisation, broadly supported

Model sheet: 27 billion parameters, dense architecture, 262,144-token context, Apache 2.0 licence. Dense model: every parameter computes on each token. The BF16 repository holds 27.8 billion elements (56 GB).

What is left to serve requests?

Qwen 3.8 27B fits comfortably on DGX Spark. In FP16, its weights take about 62.1 GB of the 128 GB available, leaving 65.9 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].

DGX Spark: indicative price €5,850 to €7,511 incl. VAT, not contractual. Source: best price €5,850 incl. VAT, €4,875 excl. VAT, listed on idealo.fr on 2 September 2026. The machine targets the TPE segment.

Where else can Qwen 3.8 27B run?

See the Qwen 3.8 27B and DGX Spark fact sheets.

How much memory does Qwen 3.8 27B need?

In FP16, weights take about 62.1 GB including the runtime margin. The attention cache sits on top and depends on context.

Does Qwen 3.8 27B fit on DGX Spark?

Qwen 3.8 27B fits comfortably on DGX Spark. In FP16, its weights take about 62.1 GB of the 128 GB available, leaving 65.9 GB for the attention cache and concurrency.

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.