🍌 Banana Serverless

Setup and host CLIP on Banana in minutes.

This repo gives a prebuilt CLIP server for use on Banana.

If you want to generalize this to deploy anything on Banana, see the guide here.

Quickstart:

The repo is already set up to run a basic CLIP model.

1. Run pip3 install -r requirements.txt to download dependencies.
2. Run python3 server.py to start the server.
3. Run python3 test.py in a different terminal session to test against it.

Make it your own:

1. Edit app.py to load and run your model.
2. Make sure to test with test.py!
3. When ready to deploy:

• edit download.py (or the Dockerfile itself) with scripts download your custom model weights at build time
• edit requirements.txt with your pip packages. Don't delete the "sanic" line, as it's a banana dependency.

Move to prod:

At this point, you have a functioning http server for your ML model. You can use it as is, or package it up with our provided Dockerfile and deploy it to your favorite container hosting provider!

If Banana is your favorite GPU hosting provider (and we sure hope it is), read on!

🍌

Deploy to Banana Serverless:

• Log in to the Banana App
• Select your customized repo for deploy!

It'll then be built from the dockerfile, optimized, then deployed on our Serverless GPU cluster and callable with any of our SDKs:

• Python
• Node JS / Typescript
• Go

You can monitor buildtime and runtime logs by clicking the logs button in the model view on the Banana Dashboard](https://app.banana.dev)

Use Banana for scale.