← Back to community benchmarks

Qwen3.6-35B-A3B-UD

M5 Pro (20c) · 64 GB · 4bit · 2026-04-18

Performance

64k

tokens

1,335

PP tok/s

35.3

TG tok/s

49086

TTFT (ms)

26.0

Peak mem (GB)

Hardware

Chip M5 Pro (20c)

Memory 64 GB

GPU Cores 20

Software

oMLX v0.3.6

macOS macOS 26.4.1

Context 65,536

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
64k	1,335	35.3	26.0 GB	current
128k	808.2	21.9	31.4 GB	view