llamaperf

Qwen3.8 27B

on M5 Pro 64GB · Inco Splash

Tone: positiveApple hardware
Sep 26, 2026
Throughput
50.0 t/s gen
Quant
UD-Q4_K_XL (GGUF)
System RAM
64 GB

Use cases

agentic

Summary

User reports Qwen3.8 27B at 50 t/s on an M5 Pro 64GB. Setup is Inco Splash with Unsloth UD-Q4_K_XL GGUF weights, run in an agentic setup. The user credits Splash's optimized kernels, speculative decoding, prefix cache and mixed-weight support, and calls it a breakthrough in local inference on Apple Silicon.