Hello everyone, I am looking into downloading the model at the deployment time so we don’t have to bake the model in the image. That way we can use same image for multiple model deployments. I got it working in local using the on_deployment hook and custom runner and when I run bentoml serve command, it runs perfectly in local. Although when I deploy the same code using the image generated by containerizing the service, it fails as it looks for model.yaml file. Of course it doesn’t exist because we are not saving the model in local bento store. Can someone help me with this if they have faced the same issue? Thanks, Mehul