Qwen3.8 27B

on AMD Strix Halo 128GB · ROCmFPX llama.cpp fork / 75e67a92b / q38rocm v1.7.0 HIP · 131,072 ctx

Oct 13, 2026

Throughput

36.8 t/s gen · 284.5 tokens/s prompt processing (prefill)Prompt processing measures input prefill speed; it does not tell you the time to first token.

Quant

Q4_0_ROCMFP4_FAST (GGUF)

KV cache

Flash Attention

MTP (Multi-Token Prediction)

System RAM

128 GB

Summary

User reports Qwen3.8-27B at 36.8 t/s generation and 284.5 t/s prefill on an AMD Strix Halo 128GB box. Setup is the ROCmFPX llama.cpp fork with Q4_0_ROCMFP4_FAST, q8_0 K and V cache, flash attention on, one parallel slot, 131072 context, and native MTP with max 4 draft tokens. The 36.8 t/s is the median of four prompt categories at temp 0, single stream with 600-token outputs; the 284.5 t/s prefill is a separate 11,626-token prompt measurement.