Deepseek Harnness - why is feels better

Guys, could someone smarter than me explain what makes Deepseek Harness so efficient? I run it with local Qwen 3.8 (Q6). I tried Opencode/Openchamber (my favourite so far), Pi agent and Hermes. New Qwen seems to overthing by default but this could be minimized with some effort. On another hand - in Deepseek Harness, it just took of - the thinking effort is correct (subjectively), it delivers great results and the content window seems to be never ending (comparing to Opencode). Not sure what is so special about DSH - is it so special, or my complicated settings for Opencode just don't work with new Qwen so well.