On an IKEA table assembly task, Fable 5.1 and Opus 5 fail, while Opus 5.5 and GPT-6 Astra both succeed.
Opus 5.5 finishes faster; GPT-6 Astra takes a cleaner, safer trajectory. Check out our full benchmark + datagen pipeline at full paper at
KatsuBluebird: The Cut He Gave Away
🌱 roots: Katsu Bluebird started smoking in late-1970s Van Nuys. By the late ’80s, he was chasing better flower in Las Vegas. Trips to Amsterdam widened his experience, but finding comparable flower back home was difficult. ☕ the name: Katsu
Claude Opus 5.5 is unbelivable
Improved my weekly review flow for my BJJ app and then made a sleek video with HyperFrames_ in less than 5 minutes to show it off. Back in the day you'd be messing around with screen record, screen studio, dragging a video into a iphone frame on
1… 2… 3… 4… 5… 6… 7… 8… 9…
10… 11… 12 TWO! Rebuilt with Opus 5.5. Enjoy!
You have reached the end of the archive
All of Claude Opus 5