Groq API - extremely fast inference with a free tier

1. Free key

Go to console.groq.com/keys -> Create API Key. The free tier is generous and does not require a card.

2. Call it with an OpenAI-compatible SDK

import OpenAI from "openai";
const client = new OpenAI({
  apiKey: process.env.GROQ_API_KEY,
  baseURL: "https://api.groq.com/openai/v1",
});
const res = await client.chat.completions.create({
  model: "llama-3.3-70b-versatile",
  messages: [{ role: "user", content: "Hi!" }],
});

Available models

• •llama-3.3-70b-versatile - default pick, strong quality
• •qwen-2.5-32b - good for code
• •mixtral-8x7b - fast MoE
• •whisper-large-v3 - cheap, fast transcription

When to use Groq vs OpenAI

Groq wins on latency (live chatbots, voice) and cost. OpenAI/Anthropic win on complex reasoning quality. Rule of thumb: prototype on Groq, promote to frontier models only where quality matters.