Qwen3.8 distilled into Qwen3.6-35B-A3B. Not official Qwen3.8-27B.
Params
35B / 3B act
Fastest decode
50 tok/s
RTX 3060 12GB · Q4_K_M GGUF · llama.cpp
Refusal
Official
Setups
1
Best reported pair · not live serving speed ↗
Runs on
Quoted one-stream decode on real boxes.
No public weights link on this desk yet.