Follow along on Twitter for detailed updates and improvements made to Banana. 45-80% Faster Cold Starts of Models When a customer calls their model and it's currently not running, a ... Read More

We did a thing today. Banana is excited to announce the launch of a new state-of-the-art computer vision model, named Carrot, developed by our (also new) open-research community, Plantain Labs.What is Plantain Labs?Plan... Read More

One of the benefits of being the provider of serverless GPU inference hosting is that we get to work with founders and teams across machine learning and hear their pain points. From these hundreds of conversations, we ide... Read More

This is a guide on building scaleable production infrastructure for ML products, and the decision points you will encounter as your product grows from zero to one million+ users.We split this article into four stages of user growth and map out the common pathways for production host... Read More

We hope you’re sitting down. Starting today, Banana is offering Serverless GPUs for Inference Hosting. There is some pretty cool stuff to unpack with this launch, but if you are pressed for time and want to get back to building here is the gist. TL;DR - Banana’s Ser... Read More