Announcement
GPT 6.1 Sol debuts at No. 3 on Macroscope’s code review leaderboard
Macroscope says it trails GPT 6 Sol on recall but leaves 30% fewer comments with very high precision.
TLDR
Macroscope says GPT 6.1 Sol debuted at No. 3 on its code review leaderboard, below GPT 6 Sol because its recall is lower. It says GPT 6.1 Sol leaves 30% fewer comments with very high precision and slightly beats Astra at one-seventh the price.
Combined views
—
2 Sources, first seen ago
