Create Evaluation Results Batch The Batch Create Evaluation Results endpoint lets you import external evaluations into the Patronus platform.
The order of input evaluations is preserved in the output.
POST / v1/ evaluation-results/ batch
X-API-KEY<token> In: header
application/jsonRequired
evaluation_resultsRequired
Evaluation Results cURL JavaScript Go Python
curl -X POST "https://api.patronus.ai/v1/evaluation-results/batch" \ -H "X-API-KEY: <token>" \ -H "Content-Type: application/json" \ -d '{ "evaluation_results": [ { "app": "string", "experiment_id": "string", "evaluator_id": "string", "profile_name": "string", "criteria": "string", "evaluated_model_system_prompt": "string", "evaluated_model_retrieved_context": [ "string" ], "evaluated_model_input": "string", "evaluated_model_output": "string", "evaluated_model_gold_answer": "string", "evaluated_model_attachments": [ { "url": "string", "media_type": "string", "usage_type": "string" } ], "pass": true, "score_raw": 0, "text_output": "string", "explanation": "string", "evaluation_duration": "string", "explanation_duration": "string", "evaluation_metadata": {}, "evaluated_model_name": "string", "evaluated_model_provider": "string", "evaluated_model_params": { "property1": 0, "property2": 0 }, "evaluated_model_selected_model": "string", "dataset_id": "string", "dataset_sample_id": 0, "tags": {} } ] }'
