GAUSS: A Workload-Aware Analytical and Monte-Carlo Simulator for LLM Inference Serving

arXiv:2610.37989v1 Announce Type: new Abstract: Large language model (LLM) inference has become an operational workload that fleets must provision, monitor, and bill against latency service-level objectives (SLOs). A single throughput number cannot answer the questions operators actually face, becau

arXiv cs.AIFreitag 05:00 · 30.10.1 Min. Lesezeit