This page covers following a Run Job's progress, reading its output, cancelling and deleting it, and how to check its result in a script.
Open the Run Jobs page and click a job to see its status, attempts and logs.
featherless-cloud runjob list
featherless-cloud runjob get <job-id>
featherless-cloud runjob logs <job-id> --follow
Output is collected while the job runs, so you can read it after the job ends. Each attempt has its own output.
featherless-cloud runjob logs <job-id>
featherless-cloud runjob logs <job-id> --attempt 2
For each attempt, the first and last 50 MB of output are kept for 30 days. Write results you need to a network volume.
Cancelling stops a job that is waiting or running. A waiting job is cancelled immediately.
Open the job's actions menu and click Cancel Run Job.
featherless-cloud runjob cancel <job-id>
Deleting removes the job and its output. A running job is stopped first.
featherless-cloud runjob delete <job-id>
exec exits with the command's own exit code, so a script can check $? after it. When the job didn't finish on its own, exec exits with 124 if it timed out, and 130 if it was cancelled or you detached from it.
If your balance reaches zero, running Run Jobs stop and fail. Waiting jobs keep waiting until you add credit. See Billing.