News for hackers. Stuff that matters.
Re:Claude for Excel by sothatsit on Monday October 27, 2025 @07:59PM (#45727810)
I also find the subjective nature of these models interesting, but in this case the difference in my experiences between Sonnet 4.5 and GPT-5 Codex, and especially GPT-5 Pro, is pretty stark. This is also not just vibes, GPT-5 consistently beats Claude models on reasoning benchmarks, often by a wide margin, and I would definitely categorise code review as a complex reasoning task (at least when looking for non-trivial bugs).

I have had GPT-5 point out dozens of complex bugs to me. Often in these cases I will try to see if other models can spot the same problems, and Gemini has occasionally but the Claude models never have (using Opus 4, 4.1, and Sonnet 4.5). Not once, which continues to surprise me. They can still be good at pointing out trivial bugs like typos, mismatching error messages, or really obvious bugs though.

If you haven’t tried it, I would try the codex /review feature and compare its results to asking Sonnet to do a review. For me, the difference is very clear for code review. For actual coding tasks, both models are much more varied, but for code review I’ve never had an instance where Claude pointed out a bug that GPT-5 missed. And I use these tools for code review all the time.

Re:Claude for Excel by bcrosby95 on Tuesday October 28, 2025 @11:15AM (#45733953)
I've noticed something similar. I've been working on some concurrency libraries for elixir and Claude constantly gets things wrong, but GPT5 can recognize the techniques I'm using and the tradeoffs.
Re:Claude for Excel by bossyTeacher on Sunday November 02, 2025 @03:31PM (#45793166)
>I have had GPT-5 point out dozens of complex bugs to me

How many of those were real and how many hallucinated?