alt Hacker NewsThey are unusable (unless you want to deliberately destroy your codebase). So if Cursor's models are Kimi based, then well. I'll skip them altogether.
Replies
jmcqk6 • today at 3:35 PM
I'm using Composer extensively, and it works great for me. Your experiences are not universal.
Bnjoroge • today at 2:56 PM
They are far from unusable. They aork great for 80-90% of a typical full stack dev. Alot less useful for more noche stuff
bel8 • today at 1:42 PM
I wouldn't skip at least testing the original. Model distilling done by Cursor could be the culprit.
esskay • today at 4:50 PM
Composer 1.x was poor. The new one is a totally different beast and absolutely fine for day to day.
ok_dad • today at 5:00 PM
I only use composer 2.5 day to day and it works fine with human review.
qingcharles • today at 3:01 PM
They're not unusable, they're just bad when compared with all the real frontier models.
Kimi works great in their CLI, but their CLI has a number of workarounds for quirks of their models, including detecting when the model gets into a loop, and reverting to a checkpoint but letting the model compose a "message" to its past self (search their CLI for "BackToTheFuture"...) It doesn't work so well in a harness that doesn't take those quirks into account.