Last updated May 16, 2025

Our Claude chat models (Claude 3.7 Sonnet, Claude 3.5 Sonnet latest, Claude 3.5 Haiku, and Claude 3.0 Haiku) generate conversational completions for input messages. This guide walks you through how to use the /v1/chat/completions API with JavaScript.

Prerequisites

1. If it’s not already installed, install the Heroku CLI. Then install the Heroku AI plugin:

   heroku plugins:install @heroku/plugin-ai
   

2. Attach a chat model to an app of yours:

   # If you don't have an app yet, you can create one with:
   heroku create $example-app # specify the name you want for your app (or skip this step to use an existing app you have)
   
   # Create and attach one of our chat models to your app:
   heroku ai:models:create -a example-app claude-3-7-sonnet --as INFERENCE
   

3. Install the necessary axios package:

   npm install axios
   

JavaScript Example Code

const axios = require('axios');

// Assert that environment variables are set
const INFERENCE_URL = process.env.INFERENCE_URL;
const INFERENCE_KEY = process.env.INFERENCE_KEY;
const INFERENCE_MODEL_ID = process.env.INFERENCE_MODEL_ID;

if (!INFERENCE_URL || !INFERENCE_KEY || !INFERENCE_MODEL_ID) {
    console.error("Missing required environment variables.");
    console.log("Set them up using the following commands:");
    console.log("export INFERENCE_URL=$(heroku config:get -a $APP_NAME INFERENCE_URL)");
    console.log("export INFERENCE_KEY=$(heroku config:get -a $APP_NAME INFERENCE_KEY)");
    console.log("export INFERENCE_MODEL_ID=$(heroku config:get -a $APP_NAME INFERENCE_MODEL_ID)");
    process.exit(1);
}

async function parseChatOutput(response) {
    if (response.status === 200) {
        console.log("Chat Completion:", response.data.choices[0].message.content);
    } else {
        console.log(`Request failed: ${response.status}, ${response.statusText}`);
    }
}

async function generateChatCompletion(payload) {
    try {
        const response = await axios.post(`${INFERENCE_URL}/v1/chat/completions`, payload, {
            headers: {
                'Authorization': `Bearer ${INFERENCE_KEY}`,
                'Content-Type': 'application/json'
            }
        });
        await parseChatOutput(response);
    } catch (error) {
        console.error("Error generating chat completion:", error.message);
    }
}

// Example payload
const payload = {
    model: INFERENCE_MODEL_ID,
    messages: [
        { role: "user", content: "Hello!" },
        { role: "assistant", content: "Hi there! How can I assist you today?" },
        { role: "user", content: "Why is Heroku so cool?" }
    ],
    temperature: 0.5,
    max_tokens: 100,
    stream: false
};

// Generate chat completion
generateChatCompletion(payload);