[frde_me]

Calling it an IDE is under-representing cursor

They have in-house models, and the data to train even more powerful ones. The cursor team is a proper AI lab.

[stymaar]

> Calling it an IDE is under-representing cursor

On the contrary, it's over selling it: it's a not even a stand-alone IDE (like Zed, for instance) it's a mere fork of VSCode.

[sebzim4500]

In the same sense that chrome is just a safari fork I suppose

[stymaar]

Safari isn't even open source, and Chrome has never been a fork of Safari.

But yes Blink definitely started as a Webkit fork, and everyone would have found that laughable if someone bought that a proprietary fork of Webkit for $60B.

[airstrike]

Isn't their in-house model just Kimi?

[frde_me]

See here https://cursor.com/blog/composer-2-5

85% of the compute for the final model is from them, and not the base Kimi model.

[airstrike]

That just means it cost a lot.

Does it perform meaningfully better than the Kimi model given all that extra compute? And proportionally to the amount spent?

[jauntywundrkind]

Cursor's Composer 2.5 is one of the few models out there focusing on coding, which is the one thing most of us here want. It's pretty good! It's not near frontier level insight generating genius, but it's regarded as very capable and trustable, and is indeed a lot better than previous Kimi. It'll be interesting to compare it versus Kimi 2.7 Code, which just dropped, which is also notably a coding specifical model. I'm expecting we'll see more of this over time and I think it has huge rewards, and Composer 2.5 is early proof.

I'm not super concerned about the spend to train the model, especially given that Kimi was famously incredibly cheaply made, and given what they are competing with. I don't think that's a meaningful concern.

Reciprocally, and in far more important relevant in my humble opinion: in terms of cost to run models: Composer 2.5 is easily one of the cheapest models out there. It's fantastically cheap. It's token efficiency is through the roof astronomical. I think this training for a coding specific model has yielded something incredibly special here, and I hope SpaceXLAIC isn't the only company doing this.

[frde_me]

That's something for us and benchmarks to decide

However it definitly isn't _just_ Kimi. The weight will be different after that 85% of extra training on top of the base model.

If those different weights are better are worse doesn't change that it's in most meaningful ways not the same as the base one.

I would encourage you to lookup their blog posts about their post training process if you want a bit more faith that they aren't running an extra 85% of compute and burning money with no-ops.

[airstrike]

"Just Kimi" is hyperbole, to be clear.

I don't think it's all no-ops. Still don't think it's a particularly relevant model/company/product.

I'll defer the reading until I see signal that they have something worthwhile. I've watched a couple interviews and used the product, neither of which impressed me.

[msdz]

You don't think that a $60b valuation is having something worthwhile?

(Only half-joking…)

[CamperBob2]

They use Kimi and post-train it on the same stuff that anyone with a Github dump can feed it. They aren't doing anything that you can't do yourself.

[redox99]

Dumping github into a model is not post training, thats pre training. And every base model already has all of github.

Composer post training is clearly very good, only second to Anthropic and OpenAI.

It does irk me a bit that they try to hide the fact that it's based on a chinese pretrained model though.

[whimsicalism]

why comment on something you clearly don't know anything about? it's on-policy RL trained not just on coding text

listen and learn :)

[tomrod]

Meh. On an outcomes analysis, I've found Cursor's delivery to be exceptionally weak.

Good luck to the alt-economy of SpaceTesla though, may all our 401ks survive.

[oompydoompy74]

Their “in house models” are reportedly basically just Kimi.

[frde_me]

Replied on the other comment about this, but putting it here:

> See here https://cursor.com/blog/composer-2-5

> 85% of the compute for the final model is from them, and not the base Kimi model.

Of course they could be lying, but it seems feasible that they are adding a lot on top of this