Quickstart

Use the OpenAI SDK or send an HTTP request to make your first Meliai API call. Choose the approach that fits your application.

Approach	Best for
OpenAI SDK	Existing Python or Node.js applications
HTTP API	Any language with an HTTP client
Integrations	LangChain, Vercel AI SDK, and other tools

Get an API key

Head to meliai.ai/account/api/keys and create a new key. Your key will look like sk-mel-.... Store it as an environment variable so you never hard-code credentials in source files:

export MELIAI_API_KEY="sk-mel-<YOUR_KEY>"

Add that line to your shell profile (.bashrc, .zshrc, etc.) so the variable persists across sessions.

Install the OpenAI SDK

Meliai is fully compatible with the official OpenAI SDK — no separate package required.

pip install openai

npm install openai

Make your first call

Point the SDK at Meliai by setting base_url to https://api.meliai.ai/v1 and supplying your Meliai API key. Everything else stays exactly the same as any OpenAI SDK usage.

import os
from openai import OpenAI

client = OpenAI(
    api_key=os.environ["MELIAI_API_KEY"],
    base_url="https://api.meliai.ai/v1",
)

response = client.chat.completions.create(
    model="<MODEL_ID>",
    messages=[{"role": "user", "content": "Name three European capitals."}],
)
print(response.choices[0].message.content)

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: process.env.MELIAI_API_KEY,
  baseURL: "https://api.meliai.ai/v1",
});

const response = await client.chat.completions.create({
  model: "<MODEL_ID>",
  messages: [{ role: "user", content: "Name three European capitals." }],
});
console.log(response.choices[0].message.content);

curl https://api.meliai.ai/v1/chat/completions \
  -H "Authorization: Bearer $MELIAI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "<MODEL_ID>",
    "messages": [{"role": "user", "content": "Name three European capitals."}]
  }'

Replace <MODEL_ID> with an actual model identifier from the Meliai catalog. You can browse models at meliai.ai/hub or fetch them programmatically with GET /v1/models. Model IDs support routing suffixes such as :speed or :eco — see the Routing page for details.

Read the response

Meliai returns a standard OpenAI-shaped response body, extended with two additional top-level fields: environment_impact and billing_cost.

{
  "id": "chatcmpl-...",
  "model": "<MODEL_ID>",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "Berlin, Paris, Madrid."
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 14,
    "completion_tokens": 8,
    "total_tokens": 22
  },
  "environment_impact": {
    "energy_kwh": 0.00015,
    "carbon_g_co2": 0.06,
    "water_liters": 0.0002,
    "renewable_percent": 85,
    "pue": 1.18,
    "provider_id": "ovhcloud",
    "location": "FR"
  },
  "billing_cost": {
    "energy": "0.0008",
    "credits": "0.0",
    "paid_with": "energy"
  }
}

The environment_impact block tells you exactly how much energy, carbon, and water this specific request consumed, which EU provider served it, and what share of the facility’s power came from renewables. The billing_cost block shows the cost broken down by energy charges and any credits applied.

Where to go next

Routing

Use :speed, :price, :eco, and :batch suffixes to control how Meliai routes your requests across providers.

Models

Browse the full catalog of 60+ open-weight models and find the right one for your task.

Streaming

Enable token-by-token streaming responses with a single extra parameter.

Migrate from OpenAI

Already using OpenAI? See the exact two-line change needed to switch.

Overview

Models & Routing

Features

Privacy

Get an API key

Install the OpenAI SDK

Make your first call

Read the response

Where to go next

Routing

Models

Streaming

Migrate from OpenAI

​Get an API key

​Install the OpenAI SDK

​Make your first call

​Read the response

​Where to go next

Routing

Models

Streaming

Migrate from OpenAI

Get an API key

Install the OpenAI SDK

Make your first call

Read the response

Where to go next