Brian Maucere

But that news just came out over from yesterday as well.

And I guess it was called Planner.

Now it's called Meet, or I'm sorry, Code Rabbit Plan.

And you can go to coderabbit.ai if you want to check that.

Yeah, so helpful.

Yeah, I'm looking at it on my screen.

It sits between, well, it sits between Opus 4.6 and GPT 5.4, which is what you said.

Although we're, you know, we're talking about on this performance calls differences here.

yeah say like i'm going to say uh 58 i'm going to say composer 2 is 62 and gbt4 is like 64. so like we're like we're really in a range now the the bigger issue here is on the x access they have cost and composer 2 is the the cheapest to use which makes a big difference too if you have a if you have a near top tier model and it's by far the cheapest

That starts to make a giant difference, especially in something like Cursor.

And again, going back to the same sort of theme we seem to be having today, which is like a lot of people are going into Cursor and they don't necessarily go in with a really strong plan or they're not using, sorry, compound engineering.

I had to go look for it in the...

If the comment, comment engineering or code rabbit plan or whatever, right?

They're not.

And so they are burning through credits while they're working on this.

And that's normal while they're sort of coming up with whatever the solution is, curse is going to help them with.

So obviously getting to lower costs.

And by the way, it's like, it's significantly cheaper.

Yeah.

I mean, Opus 4.6 High is, this is median cost per task is the way they have it labeled here.