{"id":"nvidia-dynamo","name":"NVIDIA Dynamo","maker":"NVIDIA","url":"https://developer.nvidia.com/dynamo","repo":"ai-dynamo/dynamo","summary":"A distributed inference framework that spreads serving of large models over many GPUs, with disaggregated prefill and decode.","details":"Dynamo works with engines such as vLLM, SGLang and TensorRT-LLM and provides routing, KV cache management and a frontend with an OpenAI-compatible API. It is aimed at data centre deployments and is released under the Apache licence.","open_source":true,"hosted":false,"speaks":["openai-api"],"verified":"2026-11-15","category":"runtimes","category_name":"Local and self-hosted model runtimes","page":"https://modelmoot.com/tools/nvidia-dynamo","reviewed":"2026-11-15","checked":{"at":"2026-11-15T15:07:43.409Z","address_answers":true},"note":"Descriptions are the site's own. Facts under \"checked\" were read automatically from the tool's address and public repository."}