Getting Started

InferenceLake is a data platform for collecting LLM inference logs and building specialized models. Get started by installing the SDK and setting up your first project.

Installation

Install via npm:

npm install inferencelake

Or initialize a new project:

npx inferencelake init

Quick Start

Replace your existing OpenAI calls with InferenceLake proxied models:

// Before

model: "gpt-4"

// After

model: "gpt-4/customer_support"

Configuration

Set your API key:

export INFERENCELAKE_API_KEY=your_key_here

Visit your dashboard to view analytics and manage your models.