POST
/
api
/
v1
/
agentic
/
evaluate
/
{model_id}
Evaluate Generated Data
curl --request POST \
  --url https://api.collinear.ai/api/v1/agentic/evaluate/{model_id} \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '{
  "metrics": {
    "preset": [
      "<string>"
    ],
    "custom": [
      {
        "name": "<string>",
        "definition": "<string>",
        "weight": 123
      }
    ]
  },
  "type": "<any>",
  "title": "<any>"
}'
{
  "assessment_id": "<string>",
  "results": {
    "action_completion": {
      "score": 123,
      "failed_sessions": [
        "<string>"
      ]
    }
  }
}

​
Parameters

​
Headers

  • Authorization: Bearer token for authenticating the API call
  • Content-Type: application/json

​
Request Example

curl --location 'https://api.collinear.ai/api/v1/synth_data/evaluate/{model_id}' \
--header 'Authorization: Bearer <your_token>' \
--header 'Content-Type: application/json' \
--data '{
  "space_id": "your_space_id",
  "judge_id": "your_judge_id"
  "generation_id": "uuid",
  "metrics": {
    "preset": ["action_completion", "tool_selection_quality"],
    "custom": [
      {
        "name": "conversation_flow",
        "definition": "def evaluate(conv):...",
        "weight": 0.3
      }
    ]
  },
  "thresholds": {
    "warning": 0.7,
    "critical": 0.5
  }
}'

​
Response

{
  "assessment_id": "uuid",
  "results": {
    "action_completion": {
      "score": 0.92,
      "failed_sessions": ["session_42"]
    },
    "tool_selection_quality": {
      "score": 0.85,
      "common_errors": ["wrong_parameter:account_id"]
    },
    "conversation_flow": {
      "score": 0.78,
      "suggestions": ["Add follow-up prompt detection"]
    }
  }
}

Authorizations

​
Authorization
string
header
required

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

Path Parameters

​
model_id
string<uuid>
required

Query Parameters

​
generation_id
string<uuid>
required

Body

application/json · any

The body is of type any.

Response

Successful Response

​
assessment_id
string
​
results
object