They should use one of their LLMs to get some better naming schemes - seriously ... | Hacker News

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

number6 on May 20, 2025 | parent | context | favorite | on: OpenAI Codex hands-on review

They should use one of their LLMs to get some better naming schemes - seriously LLMs are pretty good at this set of task

IanCal on May 21, 2025 [–]

They absolutely cannot be worse than the humans involved here. gpt-4o followed by a series of o models so you have gpt-4o and o4? Wonderful.

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