Skip to main content
Manage and monitor your model training runs from the command line. View run status, stream logs in real time, and poll for completion. Alias: inf train

​
List Training Runs

Display a table of training runs in the active project. 
inf training list
Alias: inf training ls

​
Options

FlagDescriptionDefault
-s, --status <status>Filter by status: queued, running, completed, failedAll statuses
-l, --limit <n>Maximum number of results20
--offset <n>Offset for pagination0
# Show only running training jobs
inf training list --status running

# Get the first 50 training runs
inf training list --limit 50
The output table includes the run ID, name, status (color-coded), base model, progress (current/total steps), loss, and creation date.

​
Get Training Run Details

View detailed information about a specific training run. 
inf training get <id>

​
Arguments

ArgumentRequiredDescription
idYesThe training job ID
Displays a detail view with all training run properties including:
  • Run ID and name
  • Status and base model
  • Current step, total steps, and loss
  • Adapter and node count
  • Eval dataset and definition (if configured)
  • Start, completion, and creation timestamps
  • Error message (if failed)

​
View Training Logs

Stream or view log entries for a training run. 
inf training logs <id>

​
Arguments

ArgumentRequiredDescription
idYesThe training job ID

​
Options

FlagDescriptionDefault
-l, --limit <n>Maximum number of log entries50
--level <level>Filter by log levelAll levels
-f, --followContinuously poll for new logs (like tail -f)Off
# View the last 50 log entries
inf training logs job_abc123

# Stream logs in real time
inf training logs job_abc123 --follow

# Only show error logs
inf training logs job_abc123 --level error
Log entries are timestamped and color-coded by level: errors in red, warnings in yellow, info in blue. In follow mode, the CLI polls for new logs every 3 seconds until you press Ctrl+C.

​
Poll Training Status

Wait for a training run to complete, printing status updates as they happen. 
inf training poll <id>

​
Arguments

ArgumentRequiredDescription
idYesThe training job ID

​
Options

FlagDescriptionDefault
-i, --interval <seconds>Poll interval in seconds10
# Poll every 10 seconds (default)
inf training poll job_abc123

# Poll every 30 seconds
inf training poll job_abc123 --interval 30
The CLI prints a status line each time the run status changes, showing the current status, progress, and loss. It exits automatically when the run reaches completed or failed. On completion, the model reference ID is displayed.