Which local model is actually good at knowing when to stop and ask you a question?

I’ve been thinking about this after using more agentic/local coding models.

A lot of the newer models are surprisingly good at continuing on their own.

But sometimes that seems like the problem.

If a requirement is ambiguous, I’d rather the model stop and ask:

“Do you mean A or B?”

instead of spending 10 minutes reasoning, making an assumption, calling tools and then confidently building the wrong thing.

I don’t see this behavior discussed much in benchmarks either. We measure coding, reasoning, tool use, context length, etc., but not really whether a model knows when it doesn’t have enough information to continue.

My genuine question for people running models locally every day is,

Which model have you found best at this?

And is it mostly the model itself, the system prompt, or your agent harness that makes the difference?