Check out the newest way to compare different models for a task/agent harness: AutoEvals

Command Reference

AutoEvals

Inspect AutoEvals analysis runs, list their eval groups, and poll runs to completion.

AutoEvals runs automated evaluations over an agent's recent traffic. Use inf autoevals to inspect an analysis run's status and outputs, list the eval run groups it launched, and poll a run to completion.

Quickstart

# Inspect a run and its eval run groups
inf autoevals get <run-id>
inf autoevals groups <run-id>

# Poll a run to completion (exits 0 on success)
inf autoevals poll <run-id>

inf autoevals get

Show one analysis run's status, stage, and outputs.

inf autoevals get <run-id>

Arguments

ArgumentRequiredDescription
runIdYesAutoEvals analysis run ID

inf autoevals groups

List a run's eval run groups and their eval runs — one group per rubric.

inf autoevals groups <run-id>

Arguments

ArgumentRequiredDescription
runIdYesAutoEvals analysis run ID

inf autoevals poll

Poll an analysis run until it reaches a terminal status (success or error).

inf autoevals poll <run-id>

Arguments

ArgumentRequiredDescription
runIdYesAutoEvals analysis run ID

Options

FlagRequiredDescriptionDefault
--interval <seconds>NoPoll interval15
--timeout <minutes>NoGive up after this many minutes45

Generated REST verbs

inf autoevals also exposes the broader AutoEvals API as generated commands — runs-list, runs-start-user, runs-cancel, insights, get-insight, usage, settings, update-settings, eval-usage-estimate, and more. Run inf autoevals --help to list them and their flags.

  • inf eval runs manual eval runs against rubrics.

On this page