llamaperf
Submit

Gemma 4 31B

on M5 Max 128GB

Tone: negativeApple hardware
May 1, 2026
Share
Throughput
7.5 t/s gen
Flash Attention
on
System RAM
128 GB

Summary

User reports Gemma4-31B at 7.5 t/s and Qwen3.6-27B locking up on an M5 Max with 128 GB, while Qwen3.6-35B-A3 is fast. Setup uses DFLASH.