3 ms·
Interesting. What model and tool was used? I have seen similar failure modes in Cursor and VSCode Copilot (using gpt5) where I have to babysit relatively small
by hu3 1y ago
Interesting. What model and tool was used?
I have seen similar failure modes in Cursor and VSCode Copilot (using gpt5) where I have to babysit relatively small refactors.
- AllegedAlec 1y agoClaude code. Whichever model it started up automatically last weekend, I didn't explicitly check.
- rglynn 1y agoThis feels like a classic Sonnet issue. From my experience, Opus or GPT-5-high are less likely to do the "narrow instruction following without making sensible wider decisions based on context" than Sonnet.
- coldtea 1y agoThis is "just use another Linux distro" all over again
- rglynn 1y agoYes and no, it's a fair criticism to some extent. Inasamuch as I would agree that different models of the same type have superficial differences. However, I also think that models which focus on higher reasoning effort in general are better at taking into account the wider context and not missing obvious implications from instructions. Non-reasoning or low-reasoning models serve a purpose, but to suggest they are akin to different flavours misses what is actually quite an important distinction.