A cost effective way of deploying models on AWS is to use 'zappa' (autotemplatin... | Hacker News

Hacker News new | past | comments | ask | show | jobs | submit

login

oli5679 on Feb 1, 2021 | parent | context | favorite | on: Show HN: Deploy ML Models on a Budget

A cost effective way of deploying models on AWS is to use 'zappa' (autotemplating for lambda + API gateway).

This just works, and costs my team less for a year than one week of Sagemaker or kubernetes architectures.

https://www.corbettanalytics.com/deploy-machine-learning-mod...

FL33TW00D on Feb 2, 2021 [–]

If you have a large model in excess of 500mb, any advice on the next best way of deploying as cost effectivtely as possible?

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact