Environments are logical divisions in your Tensorfuse runtime that enable you to deploy the same application (or a group of applications) in multiple instances for distinct purposes. For example, you can have a dev environment, a staging environment, and a production environment. Each environment can have its own set of resources, configurations, and deployments.

Tensorfuse has a default environment when it is initially configured and all your deployments without the --env flag are deployed in the default environment. New environments can be created using the following CLI command :-

tensorkube environment create --env-name <env-name>

You can list all the created environments using the following command :-

tensorkube environment list

Most CLI commands also support an --env flag that allows you to specify the environment in which you want to run the command. For example, to deploy an application in the staging environment, you can run the following command :-

tensorkube deploy --env staging

You can list deployments in a specific environment using the following command :-

tensorkube list deployments --env staging

To list deployments across all environments, you can run the following command :-

tensorkube list deployments --all
Getting Started Run SAM2 on your own AWS on Serverless GPUs