← Back to community benchmarks

Qwen3.6-35B-A3B-5bit

M4 Pro (16c) · 48 GB · 5bit · 2026-05-18

Performance

16k

tokens

644.9

PP tok/s

54.2

TG tok/s

25406

TTFT (ms)

25.8

Peak mem (GB)

Hardware

Chip M4 Pro (16c)

Memory 48 GB

GPU Cores 16

Software

oMLX v0.3.8

macOS macOS 15.7.7

Context 16,384

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
8k	674.0	59.3	24.9 GB	view
16k	644.9	54.2	25.8 GB	current
32k	565.9	47.0	27.3 GB	view
64k	431.4	36.3	30.3 GB	view