Plan, deploy & apply
Review infrastructure changes before applying them, then verify the running result.
Plan the change
python -m infimal plan imports a module and calculates the desired change for its App. The module argument is a Python import path, not a filename. If the module defines several applications, use --app to select one.
python -m infimal plan asr-emotion-checkpoint4200
python -m infimal plan services.asr-emotion-checkpoint4200 --app asr-emotion-checkpoint4200Build and place
python -m infimal deploy starts the build and placement flow. Initialization code produces the reusable snapshot; placement exposes the service in the configured environment. Ensure your account and key are authorized before running it.
python -m infimal deploy asr-emotion-checkpoint4200Apply an updated definition
python -m infimal apply reviews the plan and decides whether to rebuild or configure the live placement. Inspect the plan after every source change. The --yes option skips the interactive prompt; use it only in an already authorized workflow.
python -m infimal plan asr-emotion-checkpoint4200
python -m infimal apply asr-emotion-checkpoint4200Verify the result
A deployment command can be accepted before the service is usable. Read status, inspect logs, and confirm the expected snapshot. Test the endpoint with a representative request through the interface configured for your model.
infimal apps status asr-emotion-checkpoint4200
infimal apps snapshots asr-emotion-checkpoint4200
infimal apps logs asr-emotion-checkpoint4200 --limit 50