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

Running Kimi K2.7 Code on DGX Station

Kimi K2.7 Code has 1000 billion parameters. DGX Station offers 748 GB of HBM3e and LPDDR5X memory. This page puts the two side by side.

Short answer. Kimi K2.7 Code barely fits on DGX Station. In NVFP4, its weights take about 575 GB against 748 GB available. The 173 GB margin disappears quickly as context grows.

How much memory does Kimi K2.7 Code 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
FP162,300 GBdoes not fitfull precision, the quality reference
FP81,150 GBdoes not fitnegligible loss on most tasks
NVFP4575 GBtightBlackwell format, native FP4 compute
Q4575 GBtightblock quantisation, broadly supported

What is left to serve requests?

Kimi K2.7 Code barely fits on DGX Station. In NVFP4, its weights take about 575 GB against 748 GB available. The 173 GB margin disappears quickly as context grows. 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.

DGX Station ranges from 117 to 169 k€, indicative and not contractual. The machine targets the PME segment.

Where else can Kimi K2.7 Code run?

See the Kimi K2.7 Code and DGX Station fact sheets.

How much memory does Kimi K2.7 Code need?

In NVFP4, weights take about 575 GB including the runtime margin. The attention cache sits on top and depends on context.

Does Kimi K2.7 Code fit on DGX Station?

Kimi K2.7 Code barely fits on DGX Station. In NVFP4, its weights take about 575 GB against 748 GB available. The 173 GB margin disappears quickly as context grows.

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.