llamaperf
Submit

Qwen3.8 27B

on M5 Ultra 256GB · MLX · 32,768 ctx

Apple hardware
Oct 9, 2026
Throughput
47.6 t/s gen · 1672.8 tokens/s prompt processing (prefill)Prompt processing measures input prefill speed; it does not tell you the time to first token.
Quant
4bit (MLX)
System RAM
256 GB

Summary

User reports Qwen3.8 27B at 47.597 t/s generation and 1672.768 t/s prompt processing on an M5 Ultra 256GB Mac Studio. Setup is native mlx_lm.benchmark (mlx-lm 0.31.3 on MLX 0.32.2) with 4-bit MLX weights, 32768 prompt tokens and 128 generated tokens over 5 trials, peak memory 20.143 GB. Figures are medians from a third-party review unit; generation ranged 47.558-47.610 t/s and prompt processing 1671.556-1673.415 t/s. Speed only, no quality evaluation, and no hardware speedup ratio is claimed.