← Back to community benchmarks

Qwen3.5-122B-A10B-Text-mxfp4

M3 Max (40c) · 128 GB · 4bit · 2026-04-25

Performance

16k

tokens

1,302

PP tok/s

38.7

TG tok/s

12585

TTFT (ms)

66.0

Peak mem (GB)

Hardware

Chip M3 Max (40c)

Memory 128 GB

GPU Cores 40

Software

oMLX v0.3.7rc2

macOS macOS 26.4.1

Context 16,384

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	445.9	47.4	62.6 GB	view
4k	484.7	44.2	63.9 GB	view
16k	1,302	38.7	66.0 GB	current