Open Dashboard
Function Accuracy
DocumentationAPI Reference

Get Evaluation Results

Hand off to an LLM

Fetch evaluation results for a batch of transformations (POST).

For each requested transformation ID the response reports one of three states: a completed result, still-pending, or failed. The POST variant accepts the ID list in the request body; use the GET variant with query parameters for simpler clients.

POST
/v3/eval/results

Authorization

API Key
x-api-key<token>

Authenticate using API Key in request header

In: header

Request Body

application/json

TypeScript Definitions

Use the request body type in TypeScript.

Response Body

application/json

application/json

curl -X POST "https://api.bem.ai/v3/eval/results" \  -H "Content-Type: application/json" \  -d '{    "transformationIDs": [      "tr_01HXAB...",      "tr_01HXCD..."    ],    "evaluationVersion": "0.1.0-gemini"  }'
{
  "results": {
    "tr_01HXAB...": {
      "fieldMetrics": {
        "/invoice/number": {
          "confidenceScore": 0.97,
          "reasoning": "Matches canonical invoice number in the source document.",
          "hallucination": false,
          "relevanceScore": 1
        }
      },
      "overallConfidence": 0.97,
      "runtime": 3.42,
      "hasHallucinations": false,
      "evaluationVersion": "0.1.0-gemini",
      "createdAt": "2026-04-23T18:05:00Z"
    }
  },
  "pending": [
    {
      "transformationId": "tr_01HXCD...",
      "createdAt": "2026-04-23T18:04:55Z"
    }
  ]
}
{
  "message": "string",
  "code": 0,
  "details": {}
}

See also

Function Accuracy

Overview of the Function Accuracy API

Get Evaluation Results GET

Next Page