Sample - OpenAI API
GET/evals/{eval_id}/runs/{run_id}/output_items/{output_item_id}

Get an eval run output item

Retrieves a specific output item from an evaluation run. Use eval_id, run_id, and output_item_id to identify the evaluation, run, and output item; the response includes grader results and the evaluated sample.

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

3 parameters
eval_idstringrequired
The identifier of the evaluation containing the run.
run_idstringrequired
The identifier of the evaluation run containing the output item.
output_item_idstringrequired
The identifier of the output item to retrieve.

2 status codes
200Returns the evaluation run output item, including its identifiers, creation timestamp, status, data source item, grader results, and evaluated sample.
objectstringrequired
The type of the object. Always "eval.run.output_item".
Allowed:eval.run.output_itemDefault:eval.run.output_item
idstringrequired
Unique identifier for the evaluation run output item.
run_idstringrequired
The identifier of the evaluation run associated with this output item.
eval_idstringrequired
The identifier of the evaluation group.
created_atintegerrequired
Unix timestamp (in seconds) when the evaluation run was created.
statusstringrequired
The status of the evaluation run.
datasource_item_idintegerrequired
The identifier for the data source item.
datasource_itemobjectrequired
Details of the input data source item.
resultsarray<EvalRunOutputItemResult>required
A list of grader results for this output item.
sampleobjectrequired
A sample containing the input and output of the evaluation run.
429Returned when the request is rejected because a rate limit was exceeded.
errorobjectrequired

Error handling

A 429 is returned when the request exceeds the rate limit. Supply eval_id, run_id, and output_item_id to identify the requested evaluation run output item.