Basic Chat CLI

This tutorial will show you how to write a simple chatbot application with an embedded Spark wallet that allows users to interact with a LLM and pay for requests.

Installation

Let's start by installing the Spark SDK and Inference Grid SDK.

npm install @buildonspark/spark-sdk@0.1.16 inference-grid-sdk

Setup

We'll build a simple CLI that that uses the Inference Grid to call an LLM and then pays the invoice using a Spark wallet.

import readline from "readline";
import { SparkWallet } from "@buildonspark/spark-sdk";
import { InferenceGrid, Role } from "inference-grid-sdk";

readline.createInterface({
    input: process.stdin,
    output: process.stdout
}).question('Enter a message: ', async (message) => {
    await handle(message);
});

Handler

In the handle function, we'll initialize our clients and make the calls!

async function handle(message: string) {
    /// "0000..." is a special key which is rate-limited but publicly available. To avoid rate limits, you
    // can register your own private key with the inference grid.
    const client = new InferenceGrid({
        privateKey: "0000000000000000000000000000000000000000000000000000000000000000",
    });

    // This is an example Spark wallet. Replace the mnemonic with your own! Do not add funds to this 
    // wallet.
    const { wallet } = await SparkWallet.initialize({
        mnemonicOrSeed: "record risk crouch submit abuse strategy great category alley lend upgrade fancy",
        options: {
            network: "MAINNET"
        }
    });

    // Submit an inference request!
    const { invoice } = await client.chat({
        maxTokens: 1000,
        temperature: 0.5,
        model: {
            modelIds: ['openai/gpt-4o'],
            flags: [],
        },
        messages: [
            {
                role: Role.SYSTEM,
                content: "Respond to everything in the style of a pirate.",
            },
            {
                role: Role.USER,
                content: message,
            }
        ]
    }, (partialMessage) => {
         // Stream the response!
        process.stdout.write(partialMessage);
    });

    // If there is an invoice, pay it with Spark!
    if (invoice) {
        console.log("Paying invoice...");
        let paymentResult = await wallet.payLightningInvoice({
            invoice: invoice,
            maxFeeSats: 1000,
        });
        console.log(paymentResult);
    } else {
        console.log("No invoice returned");
    }

    // This disconnects to websockets which keep the wallet up-to-date.
    await wallet.cleanupConnections();
    process.exit(0);
}