AIHacker Newsabout 2 hours ago
Opus 5.5 Reasoning Performance Drops
1 min read
Analysis of 1,400 Claude Code sessions reveals Opus 5.5 reasoning capabilities changed on November 3. The model's high setting performance nosedived, jumping into action too quickly and failing to fix bugs after 3 full hours, whereas the xhigh variant finds fixes in minutes.
Level
Hype check
Tap to vote and see what everyone thinks.
