Serving & Orchestration
NVIDIA Dynamo
nvidiaDisaggregated-serving framework for multi-node LLM inference — splits prefill and decode across GPU pools and manages distributed KV-cache placement, tiering, and cache-aware routing on top of vLLM, TensorRT-LLM, or SGLang backends
Editorial compatibility: these components can be used together in representative deployments. That's not a requirement or a validated implementation. For real deployments, see the sourced reference architectures.