How to turn off thinking mode
#82
by
Gierry
- opened
I know there are three modes, but the Low mode is still too slow in some scenarios. Is there a way to completely disable the thinking mode like qwen's no_think?
can you test with option:
"chat_template_kwargs": {"enable_thinking": false}
during API request ?
AFAICT, there's no real way to stop the thinking, you can only control it via Reasoning Effort