node-llama-cpp
Version:
Run AI models locally on your machine with node.js bindings for llama.cpp. Enforce a JSON schema on the model output on the generation level
19 lines (18 loc) • 737 B
TypeScript
import { GbnfTerminal } from "../GbnfTerminal.js";
import { GbnfGrammarGenerator } from "../GbnfGrammarGenerator.js";
import { GbnfJsonScopeState } from "../utils/GbnfJsonScopeState.js";
export declare class GbnfArray extends GbnfTerminal {
readonly items?: GbnfTerminal;
readonly prefixItems?: GbnfTerminal[];
readonly minItems: number;
readonly maxItems?: number;
readonly scopeState: GbnfJsonScopeState;
constructor({ items, prefixItems, minItems, maxItems, scopeState }: {
items?: GbnfTerminal;
prefixItems?: GbnfTerminal[];
minItems?: number;
maxItems?: number;
scopeState: GbnfJsonScopeState;
});
getGrammar(grammarGenerator: GbnfGrammarGenerator): string;
}