User realizes the anonymous Space Bunny they tested against Opus 5 was an early MiniMax M3.1 Flash
The author had pitted the anonymous Space Bunny against Opus 5 and only noticed it was unusually fast; they now learn it was an early MiniMax M3.1 Flash, whose official release scores 62.2 on Automation-Bench ahead of Opus 5.
wait what, Space Bunny was MiniMax M3.1 Flash?? 😳 i had zero idea, i put it against Opus 5 and just thought some random stealth model was crazy fast and what i used was only an early version, the official release is even better it scores 62.2 on Automation-Bench, ahead of Opus 5, so it can handle long tasks across different tools on its own without u fixing every step 👀 cant wait to test M3.1 flash
This post quotes another post: x.com/k2sbhai/status/2103136355656344030