← Back to community benchmarks

Qwen3.6-35B-A3B

M5 (10c) · 32 GB · 4bit · 2026-07-20

Performance

64k

tokens

502.4

PP tok/s

25.7

TG tok/s

130442

TTFT (ms)

23.2

Peak mem (GB)

Hardware

Chip M5 (10c)

Memory 32 GB

GPU Cores 10

Software

oMLX v0.5.1

macOS macOS 26.5.2

Context 65,536

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	1,080	57.7	19.2 GB	view
4k	1,273	55.3	19.9 GB	view
8k	1,191	52.5	20.4 GB	view
16k	1,034	47.4	21.1 GB	view
32k	762.6	37.6	22.4 GB	view
64k	502.4	25.7	23.2 GB	current