node-llama-cpp/dist/utils/getQueuedTokensBeforeStopTrigger.d.ts

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

import { Token, Tokenizer } from "../types.js";
import { StopGenerationDetector } from "./StopGenerationDetector.js";
export declare function getQueuedTokensBeforeStopTrigger(triggeredStops: ReturnType<typeof StopGenerationDetector["prototype"]["getTriggeredStops"]>, partiallyFreeTokens: {
    tokens: Token[];
    text: string;
}, tokenizer: Tokenizer): Token[];