News for hackers. Stuff that matters.
Re:GPT-Rosalind-6 System Card Addendum by Starlevel004 on Monday November 02, 2026 @12:01PM (#50339833)
Sol 6.1 is incredibly incredibly slow, 20t/s slow (vs the ~70-80 of Terra 6.1 and Luna 6). It wouldn't be so much slower if it was just a distillation, surely? There's something else going on there.
Re:GPT-Rosalind-6 System Card Addendum by lostmsu on Monday November 02, 2026 @12:03PM (#50339873)
But is it slower than Astra?
Re:GPT-Rosalind-6 System Card Addendum by lnkdinsuxs on Monday November 02, 2026 @12:50PM (#50340671)
> Sol 6.1 is incredibly incredibly slow, 20t/s slow

How are you getting these numbers?

Re:GPT-Rosalind-6 System Card Addendum by Starlevel004 on Monday November 02, 2026 @12:59PM (#50340798)
I have a small token/s extension for Pi. It's regularly around ~15-20t/s for Sol 6.1.
Re:GPT-Rosalind-6 System Card Addendum by Silagi on Monday November 02, 2026 @01:28PM (#50341286)
There was some (admittedly flimsy) evidence from a microsoft model description card that it's doing multiple passes through its own weights before token generation to improve output; it might be an architecture change that's making it slow, moreso than raw size
Re:GPT-Rosalind-6 System Card Addendum by willy_k on Monday November 02, 2026 @03:50PM (#50343519)
Is it using comparable tokens per task?