Preserve thinking on or off? (Qwen 3.6)

r/LocalLLaMA
Open Source AI

Are y'all using the preserve thinking flag or do you have it off? If so, why? submitted by /u/My_Unbiased_Opinion [link] [comments]