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

>Gemini 2.5 Pro is the best model

It's the second time I read this in this thread. May I ask why you think this is the case? And in which domains? I am very satisfied with 2.5 pro when it comes to philosophical/literary analysis, probably because of the super long context I can fill with whole books, and wanted to try Claude Code for the same purpose, but with folders, summaries, etc to make up for the shorter context length.



I've just found it to be the best in practice, especially for more complicated debugging with code.

But also for text review on posts, etc.

Before Claude had the edge with agentic coding at least, but now even that is slipping.


I think it's at least equal best at research math.




Consider applying for YC's Fall 2025 batch! Applications are open till Aug 4

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

Search: