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

Serving DeepSeek V4 with Dynamo

DeepSeek V4 has 1600 billion parameters. Dynamo targets distributed serving across several nodes. This page gives the memory required and the platforms that qualify.

Short answer. Dynamo serves DeepSeek V4 as soon as the machine offers at least 920 GB of memory, the weight footprint in NVFP4. 4 of the 8 platforms in the catalogue meet that bar.

What is the minimum memory?

DeepSeek V4 totals 1600 billion parameters. In NVFP4 its weights take about 920 GB including the runtime margin. In FP8 the footprint doubles, to about 1,840 GB. The attention cache sits on top and depends on the context served.

What is Dynamo for?

Dynamo targets distributed serving across several nodes. See the Dynamo and DeepSeek V4 fact sheets.

On which platforms?

How much memory for DeepSeek V4 with Dynamo?

About 920 GB in NVFP4 for the weights, excluding the attention cache.

Does Dynamo suit DeepSeek V4?

Dynamo serves DeepSeek V4 as soon as the machine offers at least 920 GB of memory, the weight footprint in NVFP4. 4 of the 8 platforms in the catalogue meet that bar.

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.