Skip to content

Deploying to 'Serverless' platforms.

Since dispike is a normal AGSI application under the hood, you should have no compatibility issues using 'serverless' platforms such as

  • Amazon Lambda (using an adapter such as magnum)
  • Google Cloud App Engine (flex and standard environements)
  • Vercel
  • Heroku

Configuration and documentation can be found all over the internet (Search for FastAPI)..

Reminder

  • AGSI entry point is hosted under .referenced_application attribute for your main bot.

  • You may need to instruct Dispike to bind to 0.0.0.0!


Deploying to AWS Lambda

Resources

  • https://towardsdatascience.com/fastapi-aws-robust-api-part-1-f67ae47390f9

  • https://github.com/iwpnd/fastapi-aws-lambda-example

Deploying to Google Cloud App Engine

Resources

  • https://medium.com/analytics-vidhya/deploying-fastapi-application-in-google-app-engine-in-standard-environment-dc061d3277a

  • https://github.com/tiangolo/fastapi/issues/228

Deploying to Vercel

Resources

  • https://blog.logrocket.com/deploying-fastapi-applications-to-vercel/
  • https://github.com/benfasoli/vercel-fastapi

Deploying to Heroku

  • https://towardsdatascience.com/how-to-deploy-your-fastapi-app-on-heroku-for-free-8d4271a4ab9
  • https://towardsdatascience.com/autodeploy-fastapi-app-to-heroku-via-git-in-these-5-easy-steps-8c7958ef5d41
  • https://github.com/ms7m/dispike/pull/35#issue-607374151