Currently working with Qwen 3.5/6 35B-A3B in the lab ; learning the "quirks" ; still a ways to go.
I don't suppose there is any news on this one? I am very much enjoying your Qwen3.6-27B-Heretic-Uncensored-FINETUNE-NEO-CODE-Di-IMatrix-MAX model, but my 16 GB VRAM is just not pulling it well enough.. I believe the 35B A3B Model could potentially unlock a lot more power if the amount of active parameters were increased/doubled (5B ish, or 6B), while still keeping easily runnable.. My knowledge on these things is still too limited, but I was likely just curious what your opinion on that would be?
