← Back to community benchmarks

Qwen3.6-35B-A3B

M5 Pro (20c) · 48 GB · 8bit · 2026-05-14

Performance

16k

tokens

1,960

PP tok/s

57.7

TG tok/s

8360

TTFT (ms)

37.4

Peak mem (GB)

Hardware

Chip M5 Pro (20c)

Memory 48 GB

GPU Cores 20

Software

oMLX v0.3.8

macOS macOS 26.5

Context 16,384

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
8k	2,103	63.3	36.5 GB	view
16k	1,960	57.7	37.4 GB	current
32k	1,653	54.1	38.9 GB	view
64k	1,081	41.4	41.9 GB	view