Ollama
To use Ollama embeddings, you need to import Ollama
from llamaindex
.
import { Ollama, serviceContextFromDefaults } from "llamaindex";
const ollamaEmbedModel = new Ollama();
const serviceContext = serviceContextFromDefaults({
embedModel: ollamaEmbedModel,
});
const document = new Document({ text: essay, id_: "essay" });
const index = await VectorStoreIndex.fromDocuments([document], {
serviceContext,
});
const queryEngine = index.asQueryEngine();
const query = "What is the meaning of life?";
const results = await queryEngine.query({
query,
});