← Back to community benchmarks

Qwen3.5-122B-A10B-Text-mxfp4副本

M5 Max (40c) · 128 GB · 4bit · 2026-03-27

Performance

8k

tokens

1,560

PP tok/s

59.5

TG tok/s

5252

TTFT (ms)

64.1

Peak mem (GB)

Hardware

Chip M5 Max (40c)

Memory 128 GB

GPU Cores 40

Software

oMLX v0.2.21

macOS macOS 26.4

Context 8,192

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	1,095	66.2	62.0 GB	view
4k	1,535	62.0	63.5 GB	view
8k	1,560	59.5	64.1 GB	current
16k	1,452	53.8	65.3 GB	view