← Back to community benchmarks

Qwen3.6-35B-A3B-UD

M5 Pro (20c) · 64 GB · 4bit · 2026-04-29

Performance

32k

tokens

1,815

PP tok/s

53.5

TG tok/s

18056

TTFT (ms)

23.5

Peak mem (GB)

Hardware

Chip M5 Pro (20c)

Memory 64 GB

GPU Cores 20

Software

oMLX v0.3.8rc1

macOS macOS 26.4.1

Context 32,768

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
8k	2,217	70.8	21.5 GB	view
32k	1,815	53.5	23.5 GB	current
128k	850.5	23.3	31.6 GB	view