← Back to community benchmarks

Qwen3.6-35B-A3B

M5 (10c) · 32 GB · 4bit · 2026-04-18

Performance

64k

tokens

369.2

PP tok/s

16.9

TG tok/s

177512

TTFT (ms)

24.9

Peak mem (GB)

Hardware

Chip M5 (10c)

Memory 32 GB

GPU Cores 10

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
16k	921.0	48.0	20.9 GB	view
32k	699.5	30.4	22.2 GB	view
64k	369.2	16.9	24.9 GB	current