Nvidia Dgx Spark

Price per GB, memory bandwidth, prefill vs decode on gpt-oss-120b, clustering and software stacks: which unified-memory desktop fits your local LLM work.

A 135 GB model against a 128 GB box forces a cluster. Here is the decision order, the parallelism maths, and what the interconnect really costs per token.

From one Spark to four: the open-weight model to serve at each node count, with measured decode, prefill and context figures from Mia's AI Lab recipes.