Qwen3.8 27B
on AMD RX 7900 XT · llama.cpp · 2,048 ctx
Summary
User reports Qwen3.8-27B at around 7-9 t/s on an RX 7900 XT 20GB, sometimes lower, and asks why it is so slow compared with a desktop PC.
Setup is llama.cpp (llama-cli) with a Q4_K_M GGUF at 2048 context, -ngl 99, on a dual Xeon E5-2630 v3 server with 192GB DDR4 1866.
The speed is given only as a range, so no single figure is recorded.