AIArmin Ronacher's Thoughts and Writings1h ago

Claude Opus 4.8 and Sonnet 5 seem worse at tool calls than older

Claude Opus 4.8 and Sonnet 5 seem worse at tool calls than older models, likely due to post-training that assumes Claude Code-like harnesses as targets

Claude Opus 4.8 and Sonnet 5 seem worse at tool calls than older

A very strange Pi issue sent me down a rabbit hole over the last two days. The short version is that newer Claude models sometimes call Pi's edit tool …

Read full article

Source: Armin Ronacher's Thoughts and Writings · Opens in new tab