So my workflow is heavily skewing towards Claude-like models, in the sense that they just "do things" and don't flap about it. OpenAI models are often like "ok I did this, I could do the next thing now, should I do that thing?" I've done some experimenting and Minimax seems to be more like Claude, b
