[giancarlostoro]

Ah is that what it is? I don't use Cursor, never saw it as being relevant to me, but would not surprise me.

[schmorptron]

Cursor's composer models are finetuned kimi

[varispeed]

They 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.

[vidarh]

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.

[jmcqk6]

I'm using Composer extensively, and it works great for me. Your experiences are not universal.

[Bnjoroge]

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]

I wouldn't skip at least testing the original. Model distilling done by Cursor could be the culprit.

[AgentMasterRace]

Composer is really good, but just like any Chinese model it needs a good plan. It's cheap and fast, in 1 month of pro I used the equivalent of 500$ in API credit for it.

[esskay]

Composer 1.x was poor. The new one is a totally different beast and absolutely fine for day to day.

[qingcharles]

They're not unusable, they're just bad when compared with all the real frontier models.

[ok_dad]

I only use composer 2.5 day to day and it works fine with human review.