### Server Memory

For example, if your code loads a Machine Learning model with **1 GB in size**, when you run one process with your API, it will consume at least 1 GB of RAM. And if you start **4 processes** (4 workers), each will consume 1 GB of RAM. So in total, your API will consume **4 GB of RAM**.

And if your remote server or virtual machine only has 3 GB of RAM, trying to load more than 4 GB of RAM will cause problems. 🚨

### Traefik

<a href="https://traefik.io/" class="external-link" target="_blank">Traefik</a> é um _proxy_ reverso / _load balancer_ de alta performance. Ele pode fazer o trabalho do _"TLS Termination Proxy"_ (à parte de outros recursos).

from fastapi import FastAPI


def fake_answer_to_everything_ml_model(x: float):
    return x * 42


ml_models = {}


@asynccontextmanager
async def lifespan(app: FastAPI):
    # Load the ML model
    ml_models["answer_to_everything"] = fake_answer_to_everything_ml_model
    yield
    # Clean up the ML models and release the resources
    ml_models.clear()

You could load it at the top level of the module/file, but that would also mean that it would **load the model** even if you are just running a simple automated test, then that test would be **slow** because it would have to wait for the model to load before being able to run an independent part of the code.

    importlib.reload(tutorial001)

    client = TestClient(tutorial001.app)
    return client


@needs_pydanticv2
def test_disable_openapi(monkeypatch):
    monkeypatch.setenv("OPENAPI_URL", "")
    # Load the client after setting the env var
    client = get_client()
    response = client.get("/openapi.json")
    assert response.status_code == 404, response.text
    response = client.get("/docs")

@lru_cache
def is_mkdocs_insiders() -> bool:
    version = metadata.version("mkdocs-material")
    return "insiders" in version


def get_en_config() -> Dict[str, Any]:
    return mkdocs.utils.yaml_load(en_config_path.read_text(encoding="utf-8"))


def get_lang_paths() -> List[Path]:
    return sorted(docs_path.iterdir())


def lang_callback(lang: Optional[str]) -> Union[str, None]:
    if lang is None:

```

### Test it

Now, you should be able to go to your docs at <a href="http://127.0.0.1:8000/docs" class="external-link" target="_blank">http://127.0.0.1:8000/docs</a>, and reload the page, it will load those assets from the new CDN.

## Self-hosting JavaScript and CSS for docs

Used by FastAPI / Starlette:

* <a href="https://www.uvicorn.org" target="_blank"><code>uvicorn</code></a> - for the server that loads and serves your application.
* <a href="https://github.com/ijl/orjson" target="_blank"><code>orjson</code></a> - Required if you want to use `ORJSONResponse`.

def test_index():
    client = TestClient(app)
    with client.websocket_connect("/") as websocket:
        data = json.loads(websocket.receive_text())
        assert data == ["app", "index"]


def test_routerindex():
    client = TestClient(app)
    with client.websocket_connect("/router") as websocket:
        data = json.loads(websocket.receive_text())
        assert data == ["app", "router2", "router", "routerindex"]

            "required": True,
        },
    },
)
async def create_item(request: Request):
    raw_body = await request.body()
    try:
        data = yaml.safe_load(raw_body)
    except yaml.YAMLError:
        raise HTTPException(status_code=422, detail="Invalid YAML")
    try:
        item = Item.model_validate(data)
    except ValidationError as e: