← Back to community benchmarks

Qwen3.5-122B-A10B

M4 Max (40c) · 128 GB · 4bit · 2026-04-04

Performance

16k

tokens

510.4

PP tok/s

42.8

TG tok/s

32099

TTFT (ms)

68.9

Peak mem (GB)

Hardware

Chip M4 Max (40c)

Memory 128 GB

GPU Cores 40

Software

oMLX v0.3.2

macOS macOS 26.4

Context 16,384

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	518.5	57.4	65.6 GB	view
4k	578.0	52.9	67.0 GB	view
8k	560.7	49.8	67.7 GB	view
16k	510.4	42.8	68.9 GB	current
32k	450.7	33.1	71.2 GB	view
64k	340.1	15.3	76.1 GB	view