LLM
The LLM is responsible for reading text and generating natural language responses to queries. By default, LlamaIndex.TS uses gpt-3.5-turbo.
The LLM can be explicitly set in the ServiceContext object.
import { OpenAI, serviceContextFromDefaults } from "llamaindex";
const openaiLLM = new OpenAI({ model: "gpt-3.5-turbo", temperature: 0 });
const serviceContext = serviceContextFromDefaults({ llm: openaiLLM });