Interface LLMJudgeConfig

Configuration for LLM Judge

interface LLMJudgeConfig {
    llmProvider: AIModelProviderManager;
    modelId?: string;
    providerId?: string;
    effort?: "low" | "high" | "medium" | "xhigh" | "max";
    temperature?: number;
    systemPrompt?: string;
    errorScore?: number;
}

Properties

llmProvider modelId? providerId? effort? temperature? systemPrompt? errorScore?

Properties

llmProvider

llmProvider: AIModelProviderManager

LLM provider manager

Optional modelId

modelId?: string

Model to use for judging.

Default

the central judge resolver (gpt-5.6)

Optional providerId

providerId?: string

Provider ID. Pinning a non-openai provider requires an explicit modelId.

Optional effort

effort?: "low" | "high" | "medium" | "xhigh" | "max"

Reasoning effort. When omitted, the resolver's default (max) applies only when the resolver's model was also selected — a pinned modelId gets no injected effort (zero request-shape change).

Optional temperature

temperature?: number

Temperature for judging (lower = more consistent)

Optional systemPrompt

systemPrompt?: string

Custom system prompt for the judge

Optional errorScore

errorScore?: number

Score returned when evaluation fails (provider error, parse failure). Defaults to 0 (fail-closed) so a judge used as a safety gate does not pass on infrastructure failure — matching EmergentJudge. Set to 0.5 to restore the prior neutral-on-error behavior for non-gating quality scoring.

Settings

Member Visibility

Theme

On This Page

llmProvidermodelIdproviderIdefforttemperaturesystemPrompterrorScore