Norlen

open intelligence, by API

Open models.
Drop-in API. Fair pricing.

Chat, code, embeddings and image through an OpenAI-compatible API. Managed infrastructure, clear quotas and predictable billing — you just write the product.

Start free Read the docs

openai-compatible5 open-source modelsunlimited text

the problem

Running AI in production shouldn't require a GPU operation.

Open models are ready — and they're great. The hard part is everything else: GPU, routing, auth, quotas, billing, availability and scale.

Building and maintaining that layer becomes your distraction. Not your product. Norlen is that layer, operated by us, behind a single API.

getting started

Three steps to your first response.

If your code already talks to OpenAI, it already talks to Norlen.

01Create your account
Sign up and start on the Free plan instantly — no card. Your API key is ready in the dashboard.
02Point your code
Swap the base URL and key in any OpenAI-compatible SDK. Nothing else changes.
03Pick a plan
A flat monthly fee: unlimited text and an image quota. Upgrade when you scale.

terminal — curl

curl https://api.norlen.io/v1/chat/completions \
  -H "Authorization: Bearer $NORLEN_API_KEY" \
  -d '{ "model": "qwen3.6-35b",
        "messages": [{"role":"user","content":"Olá!"}] }'

the models

Five open-source models, one API.

Call any of them by id. Text, code, vectors and image — without switching providers.

chat · general

Qwen3.6 35B

qwen3.6-35b

The flagship. Conversation, reasoning and general-purpose tasks.

16K context

code

Qwen3 Coder

qwen3-coder

Writes, explains and reviews code across dozens of languages.

16K context

fast

Gemma 4 12B

gemma-4-12b

Light and fast for short replies. Available on the Free plan.

on Free plan

embeddings

Qwen3 Embedding

qwen3-embedding

Vectors for semantic search, RAG and classification.

2560 dimensions

image

Qwen-Image

qwen-image

Generates and edits images from text instructions.

quota per plan

reference

API documentation

Endpoints, parameters and examples per model.

Open docs

the api

One API. Any SDK.

Norlen speaks the same protocol as OpenAI. Rewrite nothing: point the client you already use at api.norlen.io/v1 and swap the key.

PythonNode.jscURLLangChainVercel AI SDK

python

from openai import OpenAI

client = OpenAI(
    base_url="https://api.norlen.io/v1",
    api_key="$NORLEN_API_KEY",
)

resp = client.chat.completions.create(
    model="qwen3.6-35b",
    messages=[{"role": "user", "content": "Olá!"}],
)
print(resp.choices[0].message.content)

why norlen

The layer that was missing between open models and your product.

Open source, without the ops

The best open models behind one API — without building GPU, routing and auth from scratch.

Fair, predictable pricing

A flat monthly fee with unlimited text and an image quota. No per-token billing, no end-of-month surprises.

Managed infrastructure

Everything runs on our infrastructure and specialized providers. Availability and scale are our problem.

OpenAI-compatible

Keep your SDK, your libs and your code. Swap the base URL and key — that's it.

plans

Fair pricing. No fine print.

Unlimited text on every paid plan. You choose by cadence and image quota — and switch anytime.

Free

To try the platform.

Free

Gemma 4 12B model
5 requests/min
Pay-as-you-go images (balance)

Start free

Starter

For individual projects.

$39/mo

Gemma 4 12B + Qwen3.6 35B + embeddings
15 requests/min
50 images/mo included
Unlimited text (fair-use)

You build the product.
Norlen delivers the intelligence.

Create free account Read the docs

free plan · no card

Open models.Drop-in API. Fair pricing.