curl --request PATCH \
--url https://app.egma.ai/v1/graders/{graderId} \
--header 'Authorization: Bearer <token>' \
--header 'Content-Type: application/json' \
--data '{}'{
"id": "<string>",
"projectId": "<string>",
"graderDefinitionId": "<string>",
"name": "<string>",
"description": "<string>",
"owner": "egma",
"type": "llm_as_judge",
"modalities": [
"chat"
],
"scopeEditable": true,
"removable": true,
"scope": {
"simulations": [
{
"kind": "all"
}
],
"production": {
"samplePercent": 50.5
}
},
"settings": {},
"passThreshold": 0.5,
"createdAt": "2023-11-07T05:31:56Z",
"updatedAt": "2023-11-07T05:31:56Z"
}{
"error": "<string>",
"message": "<string>",
"details": {}
}{
"error": "<string>",
"message": "<string>",
"details": {}
}{
"error": "<string>",
"message": "<string>",
"details": {}
}{
"error": "<string>",
"message": "<string>",
"details": {}
}{
"error": "<string>",
"message": "<string>",
"details": {}
}{
"error": "<string>",
"message": "<string>",
"details": {}
}Update a project grader's policy
Change model settings, scope, or threshold for future work. Existing simulation plans keep their selected definitions and settings, including when regraded. Expected behaviors has fixed scope; its model and threshold are editable.
curl --request PATCH \
--url https://app.egma.ai/v1/graders/{graderId} \
--header 'Authorization: Bearer <token>' \
--header 'Content-Type: application/json' \
--data '{}'{
"id": "<string>",
"projectId": "<string>",
"graderDefinitionId": "<string>",
"name": "<string>",
"description": "<string>",
"owner": "egma",
"type": "llm_as_judge",
"modalities": [
"chat"
],
"scopeEditable": true,
"removable": true,
"scope": {
"simulations": [
{
"kind": "all"
}
],
"production": {
"samplePercent": 50.5
}
},
"settings": {},
"passThreshold": 0.5,
"createdAt": "2023-11-07T05:31:56Z",
"updatedAt": "2023-11-07T05:31:56Z"
}{
"error": "<string>",
"message": "<string>",
"details": {}
}{
"error": "<string>",
"message": "<string>",
"details": {}
}{
"error": "<string>",
"message": "<string>",
"details": {}
}{
"error": "<string>",
"message": "<string>",
"details": {}
}{
"error": "<string>",
"message": "<string>",
"details": {}
}{
"error": "<string>",
"message": "<string>",
"details": {}
}Authorizations
An Egma API key.
Path Parameters
1Query Parameters
1Body
Simulation selectors and an optional production sample. Only graders with scopeEditable set to true accept scope changes.
Show child attributes
Show child attributes
Complete values for the definition's settingDefinitions. LLM graders use llm_provider and llm_model. Response latency uses maximum_response_time_ms, a positive integer in milliseconds. These settings belong to this project and do not create a definition version.
This grader passes when its score is at least this value. There is no overall run pass threshold.
0 <= x <= 1Response
The updated project grader.
111egma, project llm_as_judge, code 1chat, voice Show child attributes
Show child attributes
0 <= x <= 1