API Overview

The Norlen API follows the OpenAI-compatible standard. Any SDK or tool that talks to OpenAI works by swapping the base URL and the key — no need to rewrite your integration.

Base URLs

Resource	Method and endpoint
Chat / text	POST https://api.norlen.io/v1/chat/completions
Embeddings	POST https://api.norlen.io/v1/embeddings
Models	GET https://api.norlen.io/v1/models
Image generation	POST https://app.norlen.io/api/v1/images/generations

The image base URL is different

Text, embeddings, and models live at api.norlen.io/v1. Image generation lives at app.norlen.io/api/v1. Authentication (Bearer key) is the same for both.

Authentication

Every call uses the Authorization: Bearer YOUR_API_KEY header. See Authentication for details and best practices.

Setting up an SDK

Point the official OpenAI SDK at Norlen:

Python

from openai import OpenAI

client = OpenAI(
    base_url="https://api.norlen.io/v1",
    api_key="your-token",
)

Node.js

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api.norlen.io/v1",
  apiKey: process.env.NORLEN_API_KEY,
});

Models

Use the exact model id in your call. Full catalog in Models.

Model	id	Use
Qwen3.6 35B	qwen3.6-35b	General chat, reasoning
Qwen3 Coder	qwen3-coder	Code
Gemma 4 12B	gemma-4-12b	Fast responses (Free)
Qwen3 Embedding	qwen3-embedding	Embeddings
Qwen-Image	qwen-image	Image generation

Reference by endpoint

• Chat completions — text generation, including streaming
• Embeddings — vectors for semantic search and RAG
• Image generation — images from text
• Models — lists the models in your plan
• Errors & limits — status codes and rate limits