[vijucat]

They're great at creating test cases out of code and/or log file excerpts. They're good at run-of-the-mill tasks whose answer one can reasonably expect to find on StackOverflow. I'm using GPT-4.1 and Clause Sonnet Thinking 3.7 with vscode + GitHub Copilot