gpu functions run

Show one run, including its error and traceback if it failed

Synopsis

Shows one run of a function: status, timings, whether it started a machine, the instance it ran on, and its cost.

For a run that did not succeed it also shows the error. The error type is the name of the Python exception the function raised, with its message and the remote traceback. When GPU.ai ended the run, the type says why: Timeout, Cancelled, LaunchError (no machine could be started), MachineLost or DependencyError.

This command shows a run. It does not start one. Functions are called from Python.

gpu functions run <function> <run-id> [flags]

Examples

# Run ids are UUIDs, as printed by `gpu functions runs`
gpu functions run embed 7b2e4d1c-0a9f-4e3b-8c6d-5f4a3b2c1d0e

# Just the error of a failed run
gpu functions run embed 7b2e4d1c-0a9f-4e3b-8c6d-5f4a3b2c1d0e -o json | jq .error

Options

  -h, --help   help for run

Options inherited from parent commands

      --api-base string   API base URL (override with GPUAI_API_BASE env) (default "https://api.gpu.ai/v1")
      --debug             Enable debug logging to stderr
  -o, --output string     Output format: table|json (default table on TTY, json otherwise)

SEE ALSO

• gpu functions - Inspect and manage your functions (Python functions that run on a GPU.ai GPU)