Skip to main content

Deploy a Flask App on QuikDB — Python Deployments Made Simple

· 3 min read
Samson Ajulor
Samson Ajulor
CEO, QuikDB

Python deployment has always been annoying. Gunicorn configs, WSGI servers, Docker multi-stage builds, pip freeze vs Poetry vs pipenv. You just want your Flask app running somewhere.

QuikDB handles all of that. If your repo has a Dockerfile, we use it. If it doesn't, we generate one. Either way, your app goes from GitHub to a live URL in about 90 seconds.

Two ways to deploy Python apps​

Option A: You have a Dockerfile​

If your repo already has a Dockerfile, QuikDB uses it as-is. This is the easiest path for Flask, FastAPI, Django, or any Python framework.

A typical Flask Dockerfile looks like this:

FROM python:3.12-slim
WORKDIR /app
COPY requirements.txt .
RUN pip install --no-cache-dir -r requirements.txt
COPY . .
EXPOSE 5000
CMD ["gunicorn", "--bind", "0.0.0.0:5000", "app:app"]

QuikDB detects the Dockerfile, builds the image, and starts the container. No extra config needed.

Option B: No Dockerfile​

If you don't have a Dockerfile, QuikDB generates one based on your repo:

  • Detects requirements.txt or pyproject.toml
  • Picks the right Python version
  • Sets up the virtual environment
  • Configures the start command

You can override any of this in the deployment wizard.

Deploy step by step​

1. Connect GitHub​

Go to compute.quikdb.com, sign in, and connect your GitHub account from the Repos page.

2. Create a deployment​

Click New Deployment, pick your Flask repo, select the branch.

QuikDB scans your repo. If it finds a Dockerfile, it shows configSource: dockerfile. If not, it auto-detects your Python setup.

Review the config:

  • Build command: usually empty for Dockerfile-based deploys (the Dockerfile handles it)
  • Start command: whatever your Dockerfile's CMD is, or gunicorn app:app for auto-detected
  • Port: make sure this matches what your app listens on

3. Add environment variables​

Flask apps usually need at least:

  • FLASK_ENV or FLASK_DEBUG
  • SECRET_KEY
  • Database URLs

Add these in the config step or later in the Environment tab. Keys with SECRET, KEY, or PASSWORD in the name are auto-encrypted.

4. Deploy​

Click Deploy. QuikDB builds the Docker image on a runner node and starts your container. For Dockerfile-based deploys, build time depends on your image complexity — a slim Python image usually takes 30-60 seconds.

Your app goes live at your-app-abc123.quikdb.net.

FastAPI works the same way​

FastAPI deploys the exact same way. The only difference is your Dockerfile CMD:

CMD ["uvicorn", "main:app", "--host", "0.0.0.0", "--port", "8000"]

Make sure the port in your Dockerfile matches the port QuikDB expects (set it in the deployment config).

What you get​

  • HTTPS out of the box
  • Health checks that monitor your app
  • Logs viewable in the dashboard
  • Zero-downtime restarts on redeploy
  • Custom domains with auto-SSL

Pricing​

Same flat pricing as all QuikDB deployments:

  • Hobby ($5/mo): 3 containers, 0.5 vCPU, 512MB RAM each
  • Pro ($19/mo): 5 containers, 1 vCPU, 1GB RAM each

No per-request charges. No bandwidth surprises.

Try it​

Test repos you can fork and deploy right now:

Go to compute.quikdb.com and deploy your Python app.