Run a Python App

For now, Space only supports Python 3.8 and 3.9. Your app also needs to follow the ASGI spec (e.g. Starlette, FastAPI, Quart, etc.) or the WSGI spec (e.g. Flask, etc.). Full frameworks like Django might require more configuration which we will not cover here.

Step 1: Code configuration

For Space to run your Python app, you need to have a file called that contains the the main instance of your app, which must be called app. You can structure your app however you want, but the file must be in the root of your project. Make sure you also have a requirements.txt file that contains the dependencies of your app.


Here is an example of a simple Starlette app:

v: 0
  - name: python-app
    src: ./src/python
    engine: python3.9

from starlette.responses import PlainTextResponse

async def app(scope, receive, send):
    assert scope['type'] == 'http'
    response = PlainTextResponse('Hello, Space!')
    await response(scope, receive, send)


Here is an example of a simple FastAPI app:

from fastapi import FastAPI

app = FastAPI()

def read_root():
    return {"Hello": "Space!"}


Here is an example of a simple Flask app:

from flask import Flask

app = Flask(__name__)

def hello_world():
    return "Hello, Space!"

Step 2: Run it on Space

Make sure you have the Space CLI installed & authenticated, then run the following command:

space new

This will create a new Space project in the current directory. You can then run the following command to push your code to Space:

Important: before space push, we recommend creating a .spaceignore file to ignore files that you don’t want to be deployed to Space. For example, you should ignore .git and .venv folders.

space push

Great job! You’ve just deployed your Python app on Space. You can now access your app at the URL provided by the CLI.


  • Space only supports Python 3.8 and 3.9
  • Space only supports ASGI and WSGI apps (HTTP servers). So raw scripts won’t work.
  • Only /tmp is writable
  • SQLite, MySQL, and PostgreSQL are not supported (we recommend using Deta Base, which is built into every Space app by default)
  • Total size of deployed app is limited to 250MB (including dependencies, source code, etc). So many large dependencies like numpy or streamlit might not work, as well as larger models.

Setting up local development

We recommend using virtual environment for local development. If you have listed your app dependencies in a requirements.txt file, you can install them to your virtual environment with the following commands:

# Create a virtual environment in the current directory
python -m venv .venv
# Activate the virtual environment
source .venv/bin/activate
# Install dependencies
pip install -r requirements.txt

# If you are using FastAPI, you will also need to install uvicorn
pip install uvicorn[standard]

Then you will need to setup the dev command inside your Spacefile. Just reference the executable file from your virtual environment:

  - name: python-app
    src: ./src/python
    engine: python3.9
    dev: .venv/bin/uvicorn main:app --reload