← Back to community benchmarks

Qwen3.6-35B-A3B

M5 (10c) · 32 GB · 4bit · 2026-07-20

Performance

8k

tokens

1,207

PP tok/s

53.0

TG tok/s

6785

TTFT (ms)

20.4

Peak mem (GB)

Hardware

Chip M5 (10c)

Memory 32 GB

GPU Cores 10

Software

oMLX v0.5.1

macOS macOS 26.5.2

Context 8,192

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	1,068	57.0	19.2 GB	view
4k	1,273	54.5	19.9 GB	view
8k	1,207	53.0	20.4 GB	current
16k	1,034	47.0	21.0 GB	view
32k	739.3	38.5	22.0 GB	view
64k	494.4	26.7	23.3 GB	view