Environments

Environments represent deployment stages where your LLM configurations run. Common examples include production, staging, and development.

Overview

Each environment:

• Has a unique name
• Contains one secret for API authentication that can be rotated.
• Can serve different variant configurations than other environments

This allows you to safely test new model configurations in staging before promoting them to production.

Creating an Environment

Navigate to the Environments section in the LLMOps dashboard to create a new environment.

Give your environment a unique name (e.g., "Production", "Staging", "Development").

Environment Secrets

Each environment has associated secrets that authenticate API requests. Think of them as API keys for that specific environment.

When making an API request, pass the environment secret as the apiKey in your OpenAI client configuration. This is sent via the standard Authorization: Bearer <secret> header:

import OpenAI from 'openai';

const openai = new OpenAI({
  baseURL: 'http://localhost:3000/llmops/api/genai/v1',
  apiKey: 'your-environment-secret', // Environment secret goes here
  defaultHeaders: {
    'x-llmops-config': 'your-config-id',
  },
});

const response = await openai.chat.completions.create({
  messages: [{ role: 'user', content: 'Hello!' }],
  model: 'gpt-4o-mini',
});

Or with curl:

curl -X POST http://localhost:3000/llmops/api/genai/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer your-environment-secret" \
  -H "x-llmops-config: your-config-id" \
  -d '{"messages": [{"role": "user", "content": "Hello!"}]}'

Example Setup

A typical setup might include:

Each environment can serve different variants of the same config, allowing you to test changes safely before they reach production.