Update judge settings
Set (or clear) the workspace’s default judge model for eval runs that don’t specify one — must be a chat model from the platform catalog offered to this workspace.
Judges run on the platform’s key and are metered to the wallet, so only platform-catalog models qualify (a workspace’s own provider keys are for inference, not judging). Needs platform:write (it is a workspace setting).
Authorizations
Your workspace API key, e.g. sk_sovereign_..., sent as Authorization: Bearer <key>.
Body
Model id (e.g. "openai/gpt-4.1") to use as the default judge, or null / omitted / empty string to revert to the house default. Any non-string, non-null value → 400 'default_judge_model must be a string or null'. Must be a chat (non-embedding) model this workspace is offered, otherwise 400 '"" isn't a chat model this workspace is offered.'
Response
200 { default_judge_model: string|null, house_default: string } — the settings after the update.
The response is of type object.