@juspay/neurolink/dist/providers/llamaCpp.d.ts

The pipe layer of an AI nervous system: one interface connecting provider neurons to your application, across three inference types — generate, stream and decide. `decide` returns typed, calibrated judgements from a non-generative model (~400ms, ~$0.00002

import type { AIProviderName } from "../constants/enums.js";
import type { NeurolinkCredentials } from "../types/index.js";
import { OpenAIChatCompletionsProvider } from "./openaiChatCompletionsBase.js";
/**
 * llama.cpp Provider — direct HTTP, no AI SDK.
 *
 * Wraps a llama-server process (https://github.com/ggerganov/llama.cpp) that
 * exposes an OpenAI-compatible API at http://localhost:8080/v1 by default.
 * llama-server hosts ONE model loaded at startup; /v1/models returns just that.
 * All request/stream/tool-loop orchestration lives in
 * `OpenAIChatCompletionsProvider`; this class only declares configuration
 * and provider-specific error mapping.
 *
 * @see https://github.com/ggerganov/llama.cpp
 */
export declare class LlamaCppProvider extends OpenAIChatCompletionsProvider {
    constructor(modelName?: string, sdk?: unknown, _region?: string, credentials?: NeurolinkCredentials["llamacpp"]);
    validateConfiguration(): Promise<boolean>;
    protected getProviderName(): AIProviderName;
    protected getDefaultModel(): string;
    protected getFallbackModelName(): string;
    protected getFallbackModels(): string[];
    protected formatProviderError(error: unknown): Error;
}