MistralAI
To use MistralAI embeddings, you need to import MistralAIEmbedding
from @llamaindex/mistral
.
Installation
Section titled “Installation”npm i llamaindex @llamaindex/mistral
import { Document, Settings, VectorStoreIndex } from "llamaindex";import { MistralAIEmbedding } from "@llamaindex/mistral";
// Update Embed ModelSettings.embedModel = new MistralAIEmbedding({ apiKey: "<YOUR_API_KEY>",});
const document = new Document({ text: essay, id_: "essay" });
const index = await VectorStoreIndex.fromDocuments([document]);
const queryEngine = index.asQueryEngine();
const query = "What is the meaning of life?";
const results = await queryEngine.query({ query,});