← Back to community benchmarks

Qwen3.5-122B-A10B

M4 Max (40c) · 128 GB · 4bit · 2026-04-05

Performance

1k

tokens

494.8

PP tok/s

52.6

TG tok/s

2069

TTFT (ms)

65.6

Peak mem (GB)

Hardware

Chip M4 Max (40c)

Memory 128 GB

GPU Cores 40

Software

oMLX v0.3.4

macOS macOS 15.7.5

Context 1,024

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	494.8	52.6	65.6 GB	current
4k	561.3	50.7	66.9 GB	view
8k	553.1	47.2	67.5 GB	view
16k	535.8	41.0	68.7 GB	view
32k	463.5	30.5	71.1 GB	view
64k	357.0	1.9	76.0 GB	view
128k	245.4	1.0	85.5 GB	view