This is a heavily interactive web application, and JavaScript is required. Simple HTML interfaces are possible, but that is not what this is.
Post
Zsolt Ero
hyperknot.com
did:plc:j6xrq6ek3hodrj75s2dtxt2k
I've just read about the new math benchmark for AIs called FrontierMath and found this interesting graph on their homepage.
Basically, it's saying that Sonnet 3.5 was much better compared to o1-preview, even though this area (solving hard math problems) is the specialty area of o1-preview.
2024-11-10T13:30:02.485Z