ai_evals

read an experiment's results, one row per case

Changed on
get/w/{workspace}/ai_evals/experiments/results/{path}

Request

  • Base URL: /api (relative, and the document was published on a file host, not at the API’s origin)
  • URL: /api/w/{workspace}/ai_evals/experiments/results/{path}
  • Auth: one of:
    • HTTP bearer
    • API key in cookie token

Path parameters

workspacestring required
pathstring required

Query parameters

idstring uuid required

the experiment to read

baselinestring uuid

The experiment every column is compared against. A delta is only computed between two scores of the same scorer id, and a column the baseline was never scored with reports it rather than showing a difference.

Response

experiment results

regressedinteger required

Cells scoring lower than the baseline, across every column.

subject_current_versioninteger

The version the subject is on now. A row that ran against an earlier one describes an agent that no longer exists.

subject_deployed_hashstring

What the agent hashes to as deployed. A run of unsaved edits carrying this hash ran exactly what is deployed now — the edits were saved — so it is a run of that version rather than of edits.

Changes