← Back to community benchmarks

community--Qwen3.5-122B-A10B

M5 Max (40c) · 128 GB · 4bit · 2026-03-30

Performance

16k

tokens

1,451

PP tok/s

54.4

TG tok/s

11290

TTFT (ms)

68.9

Peak mem (GB)

Hardware

Chip M5 Max (40c)

Memory 128 GB

GPU Cores 40

Software

oMLX v0.3.0

macOS macOS 26.4

Context 16,384

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
4k	1,538	64.1	67.0 GB	view
8k	1,567	60.5	67.7 GB	view
16k	1,451	54.4	68.9 GB	current