Sample - OpenAI API
GET/evals/{eval_id}/runs

List runs for an evaluation

Lists evaluation runs for a specified evaluation. Use after to continue from the last run in a previous page, limit to control the number of runs, and status or order to filter and sort the results.

  • RetriesRetries up to 2×, 500ms backoff, 30s timeout.

5 parameters
eval_idstringrequired
The unique identifier of the evaluation whose runs to retrieve.
afterstringoptional
The identifier of the last run from the previous request, used to continue pagination.
limitintegeroptional
The number of runs to retrieve. Defaults to 20 when omitted.
Default:20
orderstringoptional
Sort order by run timestamp: `asc` or `desc`. Defaults to `asc` when omitted.
Allowed:ascdescDefault:asc
statusstringoptional
Filter runs by status: `queued`, `in_progress`, `completed`, `canceled`, or `failed`.
Allowed:queuedin_progresscompletedcanceledfailed

2 status codes
200Returns a list object containing evaluation runs, the first and last run identifiers, and a `has_more` flag indicating whether additional runs are available.
objectstringrequired
The type of this object. It is always set to "list".
Allowed:listDefault:list
dataarray<EvalRun>required
An array of eval run objects.
first_idstringrequired
The identifier of the first eval run in the data array.
last_idstringrequired
The identifier of the last eval run in the data array.
has_morebooleanrequired
Indicates whether there are more evals available.
429Returned when the request is rejected because the rate limit has been exceeded.
errorobjectrequired

Error handling

eval_id must identify the evaluation whose runs you want to list. order must be asc or desc, and status must be one of queued, in_progress, completed, canceled, or failed. A 429 is returned when the request is rejected because the rate limit has been exceeded.