← Back to community benchmarks

Qwen3.6-35B-A3B-UD

M3 Max (30c) · 36 GB · 3bit · 2026-05-03

Performance

32k

tokens

1,995

PP tok/s

25.4

TG tok/s

16424

TTFT (ms)

20.1

Peak mem (GB)

Hardware

Chip M3 Max (30c)

Memory 36 GB

GPU Cores 30

Software

oMLX v0.3.8

macOS macOS 26.4.1

Context 32,768

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	949.6	231.6	17.8 GB	view
4k	513.4	68.9	17.9 GB	view
8k	1,117	66.3	18.3 GB	view
16k	2,251	26.1	19.4 GB	view
32k	1,995	25.4	20.1 GB	current