← Back to community benchmarks

Qwen3.5-122B-A10B

M3 Ultra (60c) · 256 GB · 4bit · 2026-07-22

Performance

64k

tokens

474.8

PP tok/s

40.5

TG tok/s

138033

TTFT (ms)

69.3

Peak mem (GB)

Hardware

Chip M3 Ultra (60c)

Memory 256 GB

GPU Cores 60

Software

oMLX v0.5.3

macOS macOS 27.0

Context 65,536

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	734.4	58.2	65.7 GB	view
4k	794.6	57.0	66.8 GB	view
8k	766.9	55.2	67.0 GB	view
16k	706.0	51.9	67.3 GB	view
32k	608.3	48.1	68.0 GB	view
64k	474.8	40.5	69.3 GB	current
128k	330.0	32.2	71.6 GB	view
195k	250.6	26.0	73.3 GB	view