Build Your First RAG Server

The core power of LarkupRAG is turning your raw data into a deployable, high-performance RAG API in minutes. You can build your server using either our visual Web UI or the developer-focused CLI.

Web UI
CLI

Configure Vector Store & Embedding Model

Start by configuring your preferred Vector Store (e.g., LanceDB, Pinecone) and Embedding Model (e.g., OpenAI, Cohere) in the Configuration tab.

Ingest Your Documents

Navigate to the Data tab to upload files, paste raw text, or scrape websites. These sources will queue up as ingestion jobs.

Run Indexing

In the Index tab, click Run Indexing Job. This process will automatically chunk your text and generate vector embeddings to store in your chosen database.

Launch the Server

Finally, go to the Server tab and click Launch Server. LarkupRAG takes your entire configuration and generates a minimal, deployable Node.js server.

Initialize the Workspace

Create a new workspace project:

larkup-rag init my-rag-server

Configure Providers (Optional)

By default, the CLI uses a local LanceDB instance. If you want to configure external stores or OpenAI embeddings, run:

larkup-rag config

Add Documents

Add multiple files or URLs to your workspace:

larkup-rag add-doc --file ./docs/manual.pdf
larkup-rag add-doc --url https://example.com/docs

Index Data

Run the indexing job to chunk and embed your data:

larkup-rag index

Launch the Server

Spin up the backend in the foreground:

larkup-rag serve

The Generated Server

When you launch your server, LarkupRAG creates an optimized backend output directory . It contains everything you need and absolutely nothing you don’t:

Zero Bloat: Only dependencies for your specific Vector Store are bundled.
No Build Step: Runs directly as a Node ESM backend (node server.mjs).

SDK Usage

Because Larkup RAG exposes an OpenAI-compatible API, you can easily connect to it using your favorite AI tools and SDKs.

Vercel AI SDK
LangChain (Python)

Use the ai and @ai-sdk/openai packages to connect to your local RAG server.

import { createOpenAI } from '@ai-sdk/openai';
import { generateText } from 'ai';

const larkup = createOpenAI({
  baseURL: 'http://localhost:8080/v1',
  apiKey: 'not-needed-for-local',
});

const { text } = await generateText({
  model: larkup('rag-model'), // The model name doesn't matter for local RAG
  prompt: 'What is LarkupRAG?',
});

console.log(text);

Use langchain-openai to connect via Python.

from langchain_openai import ChatOpenAI

llm = ChatOpenAI(
    openai_api_base="http://localhost:8080/v1",
    openai_api_key="not-needed-for-local",
    model_name="rag-model"
)

response = llm.invoke("What is LarkupRAG?")
print(response.content)

Next Steps

Your server is now live at http://localhost:8080.

Deploy your server to Vercel, Docker, or a VPS.
Build a frontend chatbot using your new API.

​The Generated Server

​SDK Usage

​Next Steps

The Generated Server

SDK Usage

Next Steps