Unfortunately in xhigh reasoning effort it will burn through 2K tokens before it has even finished its bullet point overview. It really is intense and obsessive. You might need ten times more!
replyYour strategy would likely help in medium reasoning effort (because there it gets caught up in the very typical Qwen looping).
Not seen looping in the “low” reasoning effort mode.
I have been using Muse Glimmer for a few days instead of A3B. It gets the job done quicker than A3B despite being several times slower.
reply>I had to fix this on 35B A3B -- I have a proxy that just shuts it down if it gets to 2K thinking tokens and injects something like "We have thought enough, let's begin working." and it almost always finishes the turn then.
replythat is amazing, thanks for sharing.