Closed mernit closed 1 year ago
Thanks for the contribution, @mernit!
Looks like you're affiliated with Beam, so I wanted to confirm one thing about your pricing.
You indicated "No" for the "Idle Time Charged?" column. We include boot time and time between requests as "idle time". Reviewing your pricing docs, it looks like you don't charge for boot time but do charge for warm containers awaiting requests (according to the documentation for keep_warm_seconds
).
Compare that with Replicate, which only charges for prediction time, even though they keep containers warm between requests. cf also #76.
Hey @charlesfrye 👋 -- sorry for the late reply.
We actually don't charge for boot time or idle time. The keep_warm_seconds
field is an optional parameter to specify how long users want to keep their containers warm, sometimes for an infinite amount of time (essentially, this can be used to disable scale-to-zero).
If a user keeps their container running for 1 hour after their last request, we'll have to charge for that. But if this optional argument isn't used, users won't be billed for time between requests.
Thanks for getting back to me!
As I understand it, Replicate does not charge for time while containers are warm, but it also does not give users the ability to configure that setting.
It seems like this property of pricing is too complex to be squashed down to a Boolean -- at least until terminology is settled -- so I suspect it'd be better put in the notes above, perhaps with some styling to make sure it's noticed.
✅ Deploy Preview for comfy-licorice-ff5651 ready!
Toggle QR Code...
Use your smartphone camera to open QR code link.
To edit notification comments on pull requests, go to your Netlify site configuration.