After the new qwen 3.8 series I almost can't run any other local models anymore.
They all feel so bad compared to them. Local AI has always been a different kind of challenge from working with frontier cloud models. Tuning, critical workflow designs, context management, local is
Qwen 3.8-27B Q8 + DFlash Q4 — this is the best build I've done.
Everything feels so alive: the tower, the night lights, the birds, the flowers, the water, the fish, the bridge, the pool...
New MoE for 3060 12GB today?
Drop now.Qwen3.8-35B-A3B still no. 3.8 give 27B dense + Flash-Next 125B-A6B. Small A3B missing. ModelScope commit say maybe hide in cave. Until then Qwen3.6-35B-A3B still king of 12GB. Alibaba_Qwen
I'm using qwen 3.8 q8. I've used two 3090s and a context of 260k. The speed feels pretty good.
You have reached the end of the archive
All of qwen38