langchain-gigachat/dist/embeddings.d.ts

GigaChat integration for LangChain.js

import { Embeddings, type EmbeddingsParams } from "@langchain/core/embeddings";
import { GigaChat, GigaChatClientConfig } from "gigachat";
/**
 * Interface for GigachatEmbeddings parameters. Extends EmbeddingsParams and
 * defines additional parameters specific to the GigaChat embeddings class.
 */
export interface GigaChatEmbeddingsParams extends EmbeddingsParams {
    /**
     * Prefix for embeddings
     * @default {"Дано предложение, необходимо найти его парафраз \nпредложение: "}
     */
    prefixQuery?: string;
    /**
     * Use prefix or not
     * @default {false}
     */
    usePrefixQuery?: boolean;
    /**
     * The maximum number of documents to embed in a single request.
     * @default {512}
     */
    batchSize?: number;
    /**
     * Whether to strip new lines from the input text. This is recommended,
     * but may not be suitable for all use cases.
     * @default {true}
     */
    stripNewLines?: boolean;
    /** Model name to use */
    model?: string;
}
/**
 * Class for generating embeddings using the GigaChat API.
 * @example
 * ```typescript
 * // Embed a query using GigaChatEmbeddings to generate embeddings for a given text
 * const model = new GigaChatEmbeddings();
 * const res = await model.embedQuery(
 *   "What would be a good company name for a company that makes colorful socks?",
 * );
 * console.log({ res });
 *
 * ```
 */
export declare class GigaChatEmbeddings extends Embeddings implements GigaChatEmbeddingsParams {
    prefixQuery: string;
    usePrefixQuery: boolean;
    batchSize: number;
    stripNewLines: boolean;
    model: string;
    protected clientConfig: GigaChatClientConfig;
    protected _client: GigaChat;
    constructor(fields?: GigaChatEmbeddingsParams & GigaChatClientConfig);
    /**
     * Method to generate embeddings for an array of documents. Splits the
     * documents into batches and makes requests to the OpenAI API to generate
     * embeddings.
     * @param texts Array of documents to generate embeddings for.
     * @returns Promise that resolves to a 2D array of embeddings for each document.
     */
    embedDocuments(texts: string[]): Promise<number[][]>;
    /**
     * Method to generate an embedding for a single document. Calls the
     * embeddingWithRetry method with the document as the input.
     * @param text Document to generate an embedding for.
     * @returns Promise that resolves to an embedding for the document.
     */
    embedQuery(text: string): Promise<number[]>;
    /**
     * Private method to make a request to the GigaChat API to generate
     * embeddings. Handles the retry logic and returns the response from the
     * API.
     * @param input String or array of strings to embedding
     * @returns Promise that resolves to the response from the API.
     */
    protected embeddingWithRetry(input: string | Array<string>): Promise<any>;
}