Qwen3.8 27B
on NVIDIA RTX 5060 Ti 16GB · llama.cpp · 100,000 ctx
Sep 7, 2026
Summary
User reports Qwen3.8-27B at about 68 t/s generation on dual RTX 5060 Ti.
Setup is speculative decoding with a Q8 KV cache at 100k context. Vision is enabled but not used.