POST
/
evals
/
{eval_id}
/
runs
Create eval run
curl --request POST \
  --url https://api.openai.com/v1/evals/{eval_id}/runs \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '{
  "name": "<string>",
  "metadata": {},
  "data_source": {
    "type": "jsonl",
    "source": {
      "type": "file_content",
      "content": [
        {
          "item": {},
          "sample": {}
        }
      ]
    }
  }
}'
{
  "object": "eval.run",
  "id": "<string>",
  "eval_id": "<string>",
  "status": "<string>",
  "model": "<string>",
  "name": "<string>",
  "created_at": 123,
  "report_url": "<string>",
  "result_counts": {
    "total": 123,
    "errored": 123,
    "failed": 123,
    "passed": 123
  },
  "per_model_usage": [
    {
      "model_name": "<string>",
      "invocation_count": 123,
      "prompt_tokens": 123,
      "completion_tokens": 123,
      "total_tokens": 123,
      "cached_tokens": 123
    }
  ],
  "per_testing_criteria_results": [
    {
      "testing_criteria": "<string>",
      "passed": 123,
      "failed": 123
    }
  ],
  "data_source": {
    "type": "jsonl",
    "source": {
      "type": "file_content",
      "content": [
        {
          "item": {},
          "sample": {}
        }
      ]
    }
  },
  "metadata": {},
  "error": {
    "code": "<string>",
    "message": "<string>"
  }
}

Authorizations

​
Authorization
string
header
required

Bearer authentication header of the form Bearer <token>, where <token> is your auth token.

Path Parameters

​
eval_id
string
required

The ID of the evaluation to create a run for.

Body

application/json
​
data_source
object
required

Details about the run's data source. A JsonlRunDataSource object with that specifies a JSONL file that matches the eval

​
name
string

The name of the run.

​
metadata
object | null

Set of 16 key-value pairs that can be attached to an object. This can be useful for storing additional information about the object in a structured format, and querying for objects via API or the dashboard.

Keys are strings with a maximum length of 64 characters. Values are strings with a maximum length of 512 characters.

Response

Successfully created a run for the evaluation

A schema representing an evaluation run.

​
object
enum<string>
default:eval.run
required

The type of the object. Always "eval.run".

Available options:
eval.run
​
id
string
required

Unique identifier for the evaluation run.

​
eval_id
string
required

The identifier of the associated evaluation.

​
status
string
required

The status of the evaluation run.

​
model
string
required

The model that is evaluated, if applicable.

​
name
string
required

The name of the evaluation run.

​
created_at
integer
required

Unix timestamp (in seconds) when the evaluation run was created.

​
report_url
string
required

The URL to the rendered evaluation run report on the UI dashboard.

​
result_counts
object
required

Counters summarizing the outcomes of the evaluation run.

​
per_model_usage
object[]
required

Usage statistics for each model during the evaluation run.

​
per_testing_criteria_results
object[]
required

Results per testing criteria applied during the evaluation run.

​
data_source
object
required

Information about the run's data source. A JsonlRunDataSource object with that specifies a JSONL file that matches the eval

​
metadata
object | null
required

Set of 16 key-value pairs that can be attached to an object. This can be useful for storing additional information about the object in a structured format, and querying for objects via API or the dashboard.

Keys are strings with a maximum length of 64 characters. Values are strings with a maximum length of 512 characters.

​
error
object
required

An object representing an error response from the Eval API.