Hacker News Reader

Comment by edunteman | Hacker News Reader

Hey! Would love to have you try https://banana.dev (bias: I'm one of the founders). We run A100s for you and scale 0->1->n->0 on demand, so you only pay for what you use.

I'm at erik@banana.dev if you want any help with it :)

View on HN

+1 on banana.dev, I used it for a side project and deployed some custom code and it was a good experience! I liked the pricing model (lack of minimums and pay for usage instead of a "plan") and how you can package up whatever code you want.

Kind of funny how cellphones went the opposite way - we all hated "paying for usage (minutes/txts))" and now we want just the Plan.

I still use that because I rarely use the phone without wifi and don't make a lot of calls. And because it's 2023 I can change to a paid plan for a month at any time in the app, so it's the best of both worlds.

I have the cheapest plan, 5GB, and an unlimited data Android tablet for a total of $30+$15=$45 for both...

I pay my cell bill a year in advance - then I never have to worry about a bill ever.

Thanks for the +1!

Small note here: our billing is changing within the next month, to up-front payments that apply as a credit balance to your account. It still won't have minimums and you'll have the option to set up auto-refill on your balance, so it will functionally remain pay-as-you-go, but just wanted to add flavor to your comment on the pricing model.

Thanks for using us btw, you rock

Looking at your pricing, is that per seconds of GPU usage or per total seconds the app is running?

Eg. might have only a few minutes of usage in an hour and the rest of the time is spent waiting for requests. How's that billed?

https://docs.banana.dev/banana-docs/core-concepts/billing You're only billed for active replica time. Call comes in, we start a replica, it handles the request, it waits around for a 10s (configurable) idle timeout to handle any additional calls, and shuts down if no other calls to serve. The idle timeout is to prevent cold boots when not necessary, but is billed, so you can get closer to pure pay-per-call pricing by reducing idle timeout.

Looks great. It appears aimed at the inference use case rather than training, yes?

yeah, we're optimizing the infra for realtime inference, though people definitely still do run training on us, with a weights upload implemented at the end of your handler.

how were you able to tell this? still trying to understand what infra is better used for inference (say realtime image category matching) vs training (feeding a chatbot huge sums of data)

Reply on news.ycombinator.com