Quickstart

Get your first LLM request logged with Helicone in under 2 minutes using the AI Gateway.

Use the OpenAI SDK to access 100+ LLM models with automatic logging and observability.

Using a third-party SDK?

The AI Gateway is our recommended integration, you get automatic fallbacks, 100+ models through one SDK, and seamless model switching.

However, we understand you might already be deeply integrated with a third-party SDK.

We support direct integrations for OpenAI, Anthropic, Azure, and 20+ other providers. You’ll still get full observability, just without the unified API benefits.

Quick Integration

Two steps to your first logged request:

1. Set up your keys

   1. Create an account: Sign up for free
   2. Helicone API Key: Go to API Keys and generate a new key
   3. Provider Keys: Add your LLM API keys (OpenAI, Anthropic, etc.) at Provider Keys

   The Helicone key authenticates with our gateway. Provider keys let you access the actual LLMs.

2. Send your first request

   Helicone’s AI Gateway is an OpenAI-compatible, unified API with access to 100+ models, including OpenAI, Anthropic, Vertex, Groq, and more.

   Bring your own key: Go to Provider Keys to securely store API keys for the providers you want to use.

   TypeScript

   import { OpenAI } from "openai";
   
   const client = new OpenAI({
     baseURL: "https://ai-gateway.helicone.ai",
     apiKey: process.env.HELICONE_API_KEY,
   });
   
   const response = await client.chat.completions.create({
     model: "gpt-4o-mini", // Or 100+ other models
     messages: [{ role: "user", content: "Hello, world!" }],
   });
   

   Once you run this code, you’ll see your request appear in the Requests tab within seconds.

   See all supported models or configure provider routing for reliability.

What’s Next?

Now that data is flowing, explore what Helicone can do for you:

Explore The Platform

Understand how Helicone solves common LLM development challenges.

Questions?

Although we designed the docs to be as self-serving as possible, you are welcome to join our Discord or contact [e-mail] with any questions or feedback you have.