Do LLMs inference engines have a way to seed their randomness? so tho have repro... | Hacker News

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

		afiori 7 months ago \| parent \| context \| favorite \| on: I read all of Cloudflare's Claude-generated commit... Do LLMs inference engines have a way to seed their randomness? so tho have reproducible outputs with still some variance if desired?

bavell 7 months ago [–]

Yes, although it's not always exposed to the end user of LLM providers.

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