Qwen3.8 27B

Sep 18, 2026

Throughput

276.0 t/s gen · 5809.0 tokens/s prompt processing (prefill)Prompt processing measures input prefill speed; it does not tell you the time to first token.

Quant

NVFP4

VRAM reported

32 GB

Summary

User reports Qwen3.8 27B at 5,809 t/s prefill and 276 t/s decode on an AMD Radeon AI PRO R9700 32GB. Setup is a custom vLLM MXFP4 build with online NVFP4 to MXFP4 conversion, using the Unsloth Dynamic v3 NVFP4 quant. Results come from BetterBench.

View source ↗