alt Hacker NewsWhile there are a hundred different evals it seems to me that we've settled on crowning the #1 coding model as the best model. So in those terms this is still behind Opus 5.5 and Astra.
While there are a hundred different evals it seems to me that we've settled on crowning the #1 coding model as the best model. So in those terms this is still behind Opus 5.5 and Astra.
Why would you think so? From the info here it could be measurably better?