PromptEvalSuiteUpdateRequest
Replaces the named fields of a suite at the revision If-Match names (1.55.0); agent_id is fixed at create, and no field takes null, so a client that dumps unset fields as null changes nothing by accident: a judge or a simulator is replaced, never removed. A run already queued keeps the configuration it started with.
An object with 6 fields.
Fields
| Field | Type | Required | Description |
|---|---|---|---|
description | string | no | An empty string clears it. |
gate | PromptEvalGateInput or PromptEvalGateNone | no | The publish gate the suite defines (1.63.0): the expectation types and must items that must hold, with the floor and margin of the judged items, which an Agent applies once its owner names the suite. Replaced whole; none removes it (1.63.0). The owner's alone while an Agent's gate names the suite. |
judge | StageRefInput | no | The model of the llm stage that reads each transcript of a simulated scenario with its rubric and returns a verdict per item, through a Vocal task template with structured output (section 9.1). Its results are advisory until it is checked against human labels, which 1.55.0 does not define and 1.63.0 does (calibration). Null until set; a run of a simulated scenario then refuses with prompt_eval_judge_required. |
name | string | no | A display name. |
repetitions | integer | no | |
simulator | StageRefInput | no | The model of the llm stage that plays the simulated callers, through a Vocal task template (section 9.1). Null until set; a run of a simulated scenario then refuses with prompt_eval_judge_required. |
Used by
The operations whose parameters, request body or responses name this shape directly.