Model serving

Configure the Gateway provided in your Namespace, and serve large language models with NVIDIA Dynamo or KServe.

📄️ Configure Gateway

Learn how to configure authentication, routing, traffic, and rate limit policies for the Gateway provided in your Namespace.

📄️ Serve a model with NVIDIA Dynamo

Learn how to deploy GLM-5.3 with DynamoGraphDeployment and connect the serving endpoint to a Gateway.

📄️ Serve model with KServe LLMInferenceService

Learn how to serve GLM-5.3 with disaggregated Prefill and Decode by using KServe LLMInferenceService.