5 ms·
It is hard to evaluate the model performance of Composer 2.5 when Cursor's harness is so awful compared to the others on the market.
by tengbretson 2mo ago
It is hard to evaluate the model performance of Composer 2.5 when Cursor's harness is so awful compared to the others on the market.
- sergiotapia 2mo agoNot my experience at all. I've been using Cursor hardcore for about two weeks now and Composer 2.5 and it's wonderful. Now with Grok 4.5 I'm quite excited about the possibilities.
- brightball 2mo agoIn what way? I spend more of my time managing than hands on lately so I legitimately don’t know.
- seunosewa 2mo agoNot true. The only issue is cost of frontier models.
- sroussey 2mo agoYeah, it’s not great—except for debugging. It shines there.
- bdlowery 2mo agoCursors whole moat is their harness. Other people have benched opus and GPT models through Claude code, codex, and cursor, and cursor came out on top everytime because of their harness.
- AgentMasterRace 2mo agoThe harness is commonly ranked one of the best. what specifically had you hating it?
- steve1977 2mo agoCare to elaborate?