vLLM Speech Serve Use the serve action to deploy speech-processing models with vLLM on Kubernetes. Create Configure a Function with the vLLM model source URL. Run Configure and deploy the vLLM service. Use Invoke the model and list its available endpoints. Overview