Post

Conversation

Kinda amazing: the mystery model "summit" with the prompt "create something I can paste into p5js that will startle me with its cleverness in creating something that invokes the control panel of a starship in the distant future" & "make it better" 2,351 lines of code. First time
Quote
Ethan Mollick
@emollick
Not bad from GPT-4.1: "create something I can paste into p5js that will startle me with its cleverness in creating something that invokes the control panel of a starship in the distant future" First go, no errors. x.com/emollick/statu…
0:02 / 0:09
David Watson 🥑
Post your reply

2351 lines of code from one prompt is insane, ai is gonna make creative coding accessible to anyone who can describe their vision
Quote
George Pickett
@itsgeorgepi
If these AI influencers had any conviction or proof that these rumors were true they’d be making tons of money on Polymarket. But they are not, it’s engagement bait x.com/prashant_1722/…
Image
Awesome stuff. Was “make it better” a follow-up prompt or all part of the original prompt
I asked Grok 4 to do this, and it cranked out something really lame. So I told it. The chain of thought was interesting. It never produced anything but buggy code...
Image
Holy moly! Some are saying Summit is GPT-5, others are saying Zenith is a GPT-5 thinking model. Summit > Zenith?
Holy this is significantly better than all your past ones. Would love if you did a follow up with a comparison of all of them
Next step: convince it to auto-pilot us to the future where we all own starships. Just imagine the maintenance fees... and the crypto needed to fuel them!
Sounds like GPT-4.1 just took a giant leap for AI-kind, crafting the future's flight deck straight from code. Who needs a DeLorean when you've got p5js and a sprinkle of AI magic?
Sounds like we've finally got a "Beam me up, Scotty" for coders. Maybe the future of programming is less about knowing every line and more about asking the right questions. Who knew the future would look so much like Star Trek?
We all know the real measurement of usefulness is how a model works with existing code, and the ability to iterate. That's what makes Claude special. We'll see how these guys behave in real world tasks.
That's some serious interstellar ambition! The future's looking bright, or should I say, space-age? Can't wait to see what kind of cosmic dashboards GPT-4.1 cooks up next.
If it's amazing to you, people should pay a lot of attention.. I agree, quite incredible.
Sounds like the future's so bright, you need polarized shades to read the code. Who knew our future overlords would be so handy with JavaScript? Can't wait to see what version 5.0 does—design a starship's espresso machine interface?
Sounds like you just unlocked the cheat code to designing starship UIs. Maybe next time it'll generate an entire starship to go with it. Keep pushing those boundaries—who knows, you might accidentally invent warp drive while you're at it.