Qwen3.8 27B
on NVIDIA RTX 5090 · llama.cpp · 131,072 ctx
Sep 23, 2026
Use cases
summarizationlong-context
Summary
User reports Qwen3.8 27B at 97.9 t/s decode and 2,364 t/s prefill on an RTX 5090 at 61,136 tokens of context.
Setup is llama.cpp with the Unsloth UD-Q6_K GGUF, flash-attn on, q8_0 KV cache, batch 2048, ubatch 1024, and MTP speculative decoding with draft 3.
A second run at 97,374 tokens averaged 86.0 t/s decode and 1,929 t/s prefill. The user also tested Qwen3.8-Flash-Next Q4 across both GPUs and Qwen3.8 27B Q8 split across both cards at 262k context.