Serving models via a production-ready API requires a lot of work and time behind it to get it right. A project of mine I’ve been working on is a serverless model hosting solution that allows you to upload your model, and it automatically gets deployed to an auto-scaling API endpoint.
Here is a 30 second video of the deployment process.
It’s still in it’s proof of concept stage right now but the main objective we’re trying to achieve with it is that we’re trying to make model deployment as easy and as simple as possible.
It is definitely not the fastest or cheapest solution out there for larger models, but we’re planning on making it much better in the future given enough demand but for now, baby steps.
Our goal is to make the user experience for deployments as seamless as possible, while still exposing performant, reliable, and secure API endpoints for your models. Think of Vercel/Netlify but for model hosting if you will.
Nothing we have right now is permanent so if you have any feedback, questions, or suggestions, about the platform or the pricing scheme I’m happy to hear them.