← Back to community benchmarks

Qwen3.5-122B-A10B3.7bit-VL

M1 Max (32c) · 64 GB · 3bit · 2026-03-26

Performance

16k

tokens

709.9

PP tok/s

24.5

TG tok/s

23079

TTFT (ms)

55.6

Peak mem (GB)

Hardware

Chip M1 Max (32c)

Memory 64 GB

GPU Cores 32

Software

oMLX v0.2.21

macOS macOS 26.3.1

Context 16,384

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	173.8	23.0	54.5 GB	view
4k	189.9	20.9	56.0 GB	view
8k	188.3	18.9	56.6 GB	view
16k	709.9	24.5	55.6 GB	current
32k	686.8	23.8	56.4 GB	view