Introducing Phi-4-mini-flash-reasoning | Hacker News

Hacker Newsnew | past | comments | ask | show | jobs | submit

		Introducing Phi-4-mini-flash-reasoning (azure.microsoft.com)
		6 points by mikece 6 months ago \| hide \| past \| favorite \| 1 comment

incomingpain 6 months ago [–]

A ton of ~4B models lately.

Besides the obvious, lack of resources, is there any reason to run these instead of a much larger model if you have the resources?

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