> Cursor still has a lot of developer data that can assist in building a model
Their composer model is seriously good. I’ve been eyeing a cursor sub just to use it in OpenCode. They have a nice moat here.
> Cursor have decent enterprise relationships (while for xAI it is ~zero)
That has a reason. Those enterprise relationships are almost certainly going to sour at least a bit, if not for Musk‘s toxic image then for his erratic behavior.
Yeah, Composer 2 is legitimately so impressive. It is my daily driver right now both on professional and personal projects. I only find myself reaching for 5.3 Codex/GPT 5.4 when exploring a lot of technical documentation or code and for Sonnet/Opus when working on UI. Everything else is Composer.
Even if it wasn't for Musk, are these relationships really worth so much? There is a certain value in being on the approved vendor list, but it seems to me that there really isn't a lot of vendor lock-in. I think most people could switch to opencode, claude code or codex pretty easily. Maybe these relationships would be worth a lot if companies signed long-term contracts, but I doubt many did.
Just to point it out, Cursor has not made any good models themselves. Composer 2 is Kimi K2.5, and they tried to pass it as their own until people noticed that the api specified it as Kimi.
Cursor has released a technical paper [1] and several blog posts [2] describing the continued pretraining and RL they do on top of Kimi K2.5.
It is true that they were not transparent about the base model that they used until the model slug was discovered by a Twitter user via the API.
[1]: https://arxiv.org/abs/2603.24477 [2]: https://cursor.com/blog/real-time-rl-for-composer
Kimi is the base, but they've done tons of finetuning on top to produce a really good completions model.