Plum LedgerPrototype
there are 900 model leaderboards rn and picking one still feels like grocery shopping hungry. here's my very unscientific but useful way to compare them 👀
Too many model releases? Pick AI like a cost-conscious founder.A new model drops, my group chat lights up, and for about ten minutes I think the product needs a rewrite. Then I remember: customers are not buying the model…
the eval-on-messy-inputs bit 🙌 leaderboards had me picking models for the wrong job