Price an unsaved evaluation

GET
/v1/evaluations/scoring/quote

The same price for an evaluation you have not saved yet. Send the sizes of what the form holds, never the text: criteriaBytes, optionsBytes and labelsBytes (UTF-8 bytes; an option counts its value, its description and 8 bytes of punctuation, a rating label its text and 40 bytes), the model, and agentId for that agent's judgement limit. The ceiling counts them, so a longer text or more options raise quote.maxUsd.

Requires an API key with the workspace:read scope.

Authorization

apiKey
headerAuthorizationBearer <token>

A secret API key from the dashboard (Developers → API keys): vl_live_…, or vl_test_… for test mode. Each operation says which scope the key needs.

Query Parameters

model?string

The judge model. Default: gpt-4o-mini.

Length1 <= length <= 128
agentId?string

The agent the evaluation applies to; its judgement limit is used.

Formatuuid
criteriaBytes?|

The UTF-8 byte size of what the judge looks for (at most 4000 characters, so 16000 bytes).

Range0 <= value <= 16000
optionsBytes?|

The size of the options as the judge prompt writes them, at most 12 options.

Range0 <= value <= 22368
labelsBytes?|

The size of the two rating labels as the judge prompt writes them.

Range0 <= value <= 1680
sampleRate?number

The share of subjects to judge, above 0 up to 1. Default: 1.

Range0 < value <= 1

Response Body

The quote for an evaluation that is not saved yet

application/json
  1. response
quote*
Example Request
curl "https://api.vlayers.ai/v1/evaluations/scoring/quote?model=gpt-4o-mini&criteriaBytes=120&agentId=$AGENT_ID" \  -H "Authorization: Bearer $VOICELAYER_API_KEY"
Example Response
{  "quote": {    "maxUsd": 0.0123,    "typicalUsd": 0.0031,    "maxScoresPerConversation": 3,    "sampleRate": 1,    "outcomeIsFree": true  }}