nanoserver 0.1.0

Lightweight HTTP API server using Python stdlib

NanoServer

Zero-dependency HTTP API server built on Python's stdlib. Simple, fast, pure Python.

from nanoserver import APIServer

server = APIServer(host="0.0.0.0", port=8000)
server.add_route('GET', '/', lambda req: {"message": "Hello World"})
server.run()

No Flask. No FastAPI. No dependencies. Just Python.

Install

With pip:

pip install nanoserver

With uv:

uv add nanoserver

Quick Start

from nanoserver import APIServer

server = APIServer(port=8000)

# GET endpoint
def home(req):
    return {"message": "Hello World!"}

def hello(req):
    return {"message": "This is hello view."}

# POST endpoint
def echo(req, data):
    return {"received": data}

server.add_route('GET', '/', home)
server.add_route('GET', '/hello', hello)
server.add_route('POST', '/echo', echo)

server.run()

Test it:

curl http://localhost:8000/hello
curl -X POST http://localhost:8000/echo -d '{"test":"data"}' -H "Content-Type: application/json"

Other Usage Examples

Simple API

from nanoserver import APIServer

server = APIServer(port=8000)

def get_status(req):
    return {"status": "running", "version": "1.0"}

def create_item(req, data):
    return {"created": data, "id": 123}

server.add_route('GET', '/status', get_status)
server.add_route('POST', '/items', create_item)

server.run()

Webhook Receiver

from nanoserver import APIServer

server = APIServer(port=9000)

def handle_webhook(req, data):
    print(f"Received: {data}")
    # Process your webhook
    return {"status": "received"}

server.add_route('POST', '/webhook', handle_webhook)
server.run()

Mock API for Testing

from nanoserver import APIServer

server = APIServer(port=3000)

def get_user(req):
    return {"id": 1, "name": "John", "email": "[e-mail]"}

def login(req, data):
    if data.get('password') == 'secret':
        return {"token": "abc123"}
    return {"error": "Invalid"}

server.add_route('GET', '/user', get_user)
server.add_route('POST', '/login', login)

server.run()

Inference API for Local LLM

from nanoserver import APIServer
from your_local_model_loader import model

server = APIServer(port=3000)


def inference(req, data):
    prompt = data.get('prompt')
    temperature = data.get('temperature')
    max_tokens = data.get('max_tokens')
    try:
        completion = model.create_chat_completion(
                prompt=prompt,
                temperature=temperature,
                max_tokens=max_tokens,
            )
        
        return completion
    except Exception as exc:
        raise exc

server.add_route('POST', '/inference', inference)

server.run()

Why NanoServer?

• 🪶 Lightweight - Pure Python stdlib, no dependencies
• ⚡ Fast - Running in miliseconds
• 🎯 Simple - One class, two methods, done
• 🔧 Flexible - Microservices, IoT, prototypes, webhooks

Perfect for quick APIs, prototypes, testing, and environments where you can't install heavy frameworks.

API

APIServer(host='0.0.0.0', port=8000)

Create server instance.

server.add_route(method, path, handler)

Register a route. method is 'GET' or 'POST'.

server.run()

Start the server.

---

Contributing

Found a bug? Want a feature? PRs welcome!

Made with ❤️ for developers who love simplicity (by Taufiq)