Earlier quoted context omitted.
It's by default so you use all those tasty tokens. Kinda wish there was a deterministic, mostly terse, language to interact with computers
Kinda, more output tokens usually correlates with better benchmark scores. Ideally LLMs would keep that in their thinking section, then draft a response (what they write currently), then output something short. It'd consume even more tokens, but we wouldn't see that text
When you see a thinking summary like "Now writing the function..."; the raw thinking is actually writing the function in its internal thinking. Occasionally, the summariser misses and you get to see the raw text from models like Opus.
You can also try an open weight LLM like Qwen3.6 and see something that probably resembles the shape of frontier model thinking in some loose way.