← Back to community benchmarks

Qwen3.6-35B-A3B-UD

M4 Pro (20c) · 64 GB · 3bit · 2026-04-17

Performance

64k

tokens

512.5

PP tok/s

39.8

TG tok/s

127871

TTFT (ms)

22.2

Peak mem (GB)

Hardware

Chip M4 Pro (20c)

Memory 64 GB

GPU Cores 20

Software

oMLX v0.3.6

macOS macOS 26.3

Context 65,536

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	755.2	68.6	16.4 GB	view
4k	895.8	66.1	17.2 GB	view
8k	881.9	63.4	17.6 GB	view
16k	770.6	58.4	18.2 GB	view
32k	636.4	49.2	19.5 GB	view
64k	512.5	39.8	22.2 GB	current