​
What is tfy apply?

tfy apply is a CLI command that lets you create or update resources, such as deployments, workspaces, clusters, and provider accounts, by applying a manifest file.

A manifest is a YAML file that describes the resource you want to create or update.

Note: If you haven’t set up the CLI yet, see the CLI setup guide.

​
Example: Deploying an OpenAI Provider Account

Here’s a sample manifest for creating an OpenAI provider account.

name: openai-account-cli
type: provider-account/openai
collaborators:
  - subject: user:gowtham.v@truefoundry.com
    role_id: provider-account-manager
  - subject: team:everyone
    role_id: provider-account-access
auth_data:
  api_key: tfy-secret://truefoundry:openai-group:API_KEY
  type: api-key
integrations:
  - type: integration/model/openai
    name: gpt-4
    model_id: gpt-4
    model_types:
      - chat
    cost:
      value:
        input: 0.03
        output: 0.06
      metric: per_1000_tokens
  - type: integration/model/openai
    name: gpt-4o
    model_id: gpt-4o
    model_types:
      - chat
    cost:
      value:
        input: 0.0025
        output: 0.01
      metric: per_1000_tokens

​
What is Dry Run?

The --dry-run flag simulates the apply process without making any changes. This is useful to validate your manifest and see what would happen if you actually applied it.

​
What is Show Diff?

The --show-diff flag, when used with --dry-run, displays the differences between your manifest and the currently deployed resources. This helps you review what changes will be made before applying them.

​
Previewing Changes Before Applying

To see what changes will be made without actually applying them, you can use the following command:

tfy apply -f <your-manifest-file>.yaml --dry-run --show-diff

This will show you a diff of what will change, without updating anything.

Diff output for dry run

​
Applying the changes

To apply the changes in your manifest, run:

tfy apply -f <your-manifest-file>.yaml

This will actually create or update the resource as described in your YAML file.