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.
What a crazy week in AI!
🚀 Ox Alpha GLM 5.3 Qwen 3.8 Flash Next Tencent Hy4 Minimax FastH3 World Humanoid Games Gemini 3.5 Transcribe Gemini Omni 1.1 Flash Block 3D One Video One World FixAnything Google PPE Code World Model VoiceMem Orbit++ DiffusionOPSD &
The model underneath it is Qwen 3.8 Max.
Alibaba says it has: 2.4T total parameters ~95B active parameters per task 1M-token context The idea is simple: Huge model capacity without activating the entire model for every job.
You have reached the end of the archive
All of qwen38