How are you grading these? Are you going on feeling, or do you have a formalized... | Hacker News

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

		_1 on March 12, 2025 \| parent \| context \| favorite \| on: Gemma3 – The current strongest model that fits on ... How are you grading these? Are you going on feeling, or do you have a formalized benchmarking process?

archerx on March 12, 2025 [–]

From just using them a lot and getting the results that I want without going "ugh!".

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