‹ BackHN Continuity

Thread

GPT 6.1 Sol: Near-Astra intelligence for a fifth of the price

1066 points · 953 comments · crorella

  1. amelius · · focus · HN ↗
    If these models are so smart, can't _they_ select the right model for each task?
    1. mholm · · focus · HN ↗
      Switching models is _very_ expensive in compute (you have to rerun everything from the beginning), and highly variable in cost. Cursor tried doing this for awhile, but inconsistent performance/usage means most users turned it off and pick models specifically.
Open on Hacker News to reply ↗

Unofficial Hacker News client; not affiliated with Y Combinator.