Skip to main content
GET
/
evaluation
/
{id}
Get a specific evaluation result by ID
curl --request GET \
  --url https://beta.getplum.ai/v1/evaluation/{id} \
  --header 'Authorization: <api-key>'
{
  "results_id": "<string>",
  "created_at": "<string>",
  "dataset_id": "<string>",
  "metrics_id": "<string>",
  "metrics_definitions": [
    "<string>"
  ],
  "pair_count": 123,
  "system_prompt": "<string>",
  "score_means": [
    123
  ],
  "score_medians": [
    123
  ],
  "score_mins": [
    123
  ],
  "score_maxes": [
    123
  ],
  "score_std_devs": [
    123
  ],
  "score_confidence_intervals": [
    {
      "ci_low": 123,
      "ci_high": 123,
      "ci_confidence": 123
    }
  ],
  "min_scoring_pairs": [
    [
      {
        "reason": "<string>",
        "pair_id": "<string>",
        "score": 123
      }
    ]
  ],
  "all_scored_pairs": [
    [
      {
        "reason": "<string>",
        "pair_id": "<string>",
        "score": 123
      }
    ]
  ],
  "human_critique": [
    {
      "id": "<string>",
      "pair_id": "<string>",
      "metric_idx": 123,
      "comment": "<string>",
      "vote": -1,
      "user": "<string>",
      "user_email": "<string>",
      "time": "<string>"
    }
  ]
}

Authorizations

Authorization
string
header
required

Path Parameters

id
string
required

Query Parameters

includeAllScores
boolean
default:false

Whether to include all scored pairs with their scores and reasons in the API response (useful for detailed analysis)

Response

200 - application/json

Successfully retrieved evaluation result

results_id
string
created_at
string
dataset_id
string
metrics_id
string
metrics_definitions
string[]
pair_count
integer
system_prompt
string
score_means
number[]
score_medians
number[]
score_mins
number[]
score_maxes
number[]
score_std_devs
number[]
score_confidence_intervals
object[]
min_scoring_pairs
object[][]
all_scored_pairs
object[][]

All scored pairs with scores and reasons (only included when includeAllScores=true)

human_critique
object[]
I