← Back to community benchmarks

Qwen3.5-122B-A10B

M3 Max (30c) · 96 GB · 4bit · 2026-04-04

Performance

16k

tokens

904.6

PP tok/s

8.8

TG tok/s

18112

TTFT (ms)

70.2

Peak mem (GB)

Hardware

Chip M3 Max (30c)

Memory 96 GB

GPU Cores 30

Software

oMLX v0.3.2

macOS macOS 26.4

Context 16,384

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	303.9	33.3	67.0 GB	view
4k	332.3	29.1	68.8 GB	view
8k	324.7	28.2	69.4 GB	view
16k	904.6	8.8	70.2 GB	current
32k	889.7	9.0	71.6 GB	view
64k	809.8	9.1	73.4 GB	view
128k	636.8	8.8	76.9 GB	view