Getting Started

Welcome, connecting your application to InferenceLake is a super simple 2 step process.

Step 1 - Setting API keys and base URL

Vercel AI SDK (coming soon)

from openai import OpenAI

client = OpenAI(
    base_url="https://api.inferencelake.com/openai/",         < update base URL
    api_key="YOUR_OPENAI_KEY" + "YOUR_INFERENCELAKE_KEY"  < update API key
)

Step 2 - Updating model calls with a model subtype

Vercel AI SDK (coming soon)

response = client.chat.completions.create(
    model="gpt-4/customer-support",    < add model subtype
    messages=[{"role": "user", "content": "Hello!"}]
)

And that's it! Your model calls are now routed through the InferenceLake network, calls will be added to your InferenceLake database and will be visible on your dashboard here.