← Back to community benchmarks

Qwen3.6-35B-A3B

M5 (10c) · 32 GB · 4bit · 2026-07-20

Performance

32k

tokens

747.1

PP tok/s

37.7

TG tok/s

43858

TTFT (ms)

22.3

Peak mem (GB)

Hardware

Chip M5 (10c)

Memory 32 GB

GPU Cores 10

Software

oMLX v0.5.1

macOS macOS 26.5.2

Context 32,768

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	1,064	55.5	19.2 GB	view
4k	1,257	55.6	19.9 GB	view
8k	1,195	53.2	20.3 GB	view
16k	1,033	42.4	21.0 GB	view
32k	747.1	37.7	22.3 GB	current
64k	458.9	24.4	22.9 GB	view