Get the current status and progress of an evaluation.
This endpoint retrieves the status of a running or completed evaluation with detailed progress metrics. Use this to track progress and determine when results are ready.
Path Parameters:
evaluation_id: Unique evaluation identifierReturns:
evaluation_id: The evaluation identifierstatus: Current status (pending, completed, failed)benchmark_id: The benchmark being used for evaluationprogress (optional): Progress string (e.g., "3/10 questions")questions_completed (optional): Number of questions completed so fartotal_questions (optional): Total number of questions in the benchmarkavg_time_per_question (optional): Average time per question in secondseta_seconds (optional): Estimated time remaining in secondsRaises:
404: If evaluation not found or doesn’t belong to the authenticated userExample Request:
GET /api/v3/evaluations/eval-xyz789/status
Headers: {"Authorization": "Bearer <api_key>"}
Example Response (Pending):
{
"evaluation_id": "eval-xyz789",
"status": "pending",
"benchmark_id": "task-abc123",
"progress": "0/10 questions",
"questions_completed": 0,
"total_questions": 10
}
Example Response (In Progress):
{
"evaluation_id": "eval-xyz789",
"status": "pending",
"benchmark_id": "task-abc123",
"progress": "5/10 questions",
"questions_completed": 5,
"total_questions": 10,
"avg_time_per_question": 12.5,
"eta_seconds": 62
}
Example Response (Completed):
{
"evaluation_id": "eval-xyz789",
"status": "completed",
"benchmark_id": "task-abc123",
"progress": "10/10 questions",
"questions_completed": 10,
"total_questions": 10,
"avg_time_per_question": 11.8
}
Notes:
completed, use /evaluations/{evaluation_id}/results to get full resultsfailed indicates an error occurred during evaluationBearer authentication header of the form Bearer <token>, where <token> is your auth token.
Returns the current status and progress of an evaluation. This endpoint retrieves the status of a running or completed evaluation with detailed progress metrics, allowing users to track the evaluation process and determine when results are ready.
Unique identifier for the evaluation
"eval-abc123"
Current status (pending, completed, failed)
"pending"
Benchmark ID used
"benchmark-abc123"
Progress string (e.g., '3/10 questions')
"3/10 questions"
Number of questions completed
3
Total number of questions
10
Average time per question in seconds
15.5
Estimated time remaining in seconds
300
Timestamp when evaluation started
"2024-02-24T10:00:00Z"
Timestamp when evaluation completed
"2024-02-24T10:05:00Z"