… Claude Code suffered three issues over the last month It was a rough period for the AI agent Over on the Anthropic website, the company published a blog post titled " An update on recent Claude Code quality reports ." In it, the company reveals three areas that affected Claude's performance and cau… …
… But only five days later, on March 4, the same company quietly dropped Claude Code's default reasoning effort from high to medium and didn't tell anyone for weeks. …
… I'm not here to argue against Claude's models They're good, end of story Running local LLMs to offset my use of Claude was a smart play, but I can't deny that Claude's models are fantastic. Opus has been the best for reasoning among any LLMs I've used, even the frontier models from other companies. …
… Opus 4.7 offers five: low, medium, high, xhigh, and max. Cherny says he uses xhigh for most tasks and only bumps it to max for the hardest ones. Xhigh is now the default in Claude Code, and you can change it anytime using the /effort slash command. …
… Introducing the five activity sheets that Claude Code will try to solve They range from easy to difficult First, let's lay out the challenge Claude Code will be performing. I will set Claude Code to Sonnet 4.6, with Medium reasoning. …
… You can use the /effort command and change how much thinking budget Claude allocates low, medium, or high to a problem before it starts spitting out code. …
… Related I tested Claude’s new interactive visuals, and they’re changing how I explain things Most LLMs suffer with visualisation, Claude doesn't It can make your schematic diagrams come alive Seeing everything in action makes everything make sense Another one of the issues I usually have with textb… …
