← Back to community benchmarks

Qwen3.6-35B-A3B-oQ6

M3 Max (40c) · 128 GB · 6bit · 2026-04-23

Performance

64k

tokens

1,880

PP tok/s

28.8

TG tok/s

34867

TTFT (ms)

52.0

Peak mem (GB)

Hardware

Chip M3 Max (40c)

Memory 128 GB

GPU Cores 40

Software

oMLX v0.3.7

macOS macOS 26.4.1

Context 65,536

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	1,160	248.7	47.9 GB	view
4k	290.5	61.3	47.9 GB	view
8k	1,373	58.3	48.3 GB	view
16k	2,357	29.9	49.4 GB	view
32k	2,329	30.2	50.0 GB	view
64k	1,880	28.8	52.0 GB	current