← Back to community benchmarks

unsloth--Qwen3.6-35B-A3B-UD

M5 Pro (20c) · 64 GB · 4bit · 2026-07-21

Performance

4k

tokens

2,457

PP tok/s

72.3

TG tok/s

1667

TTFT (ms)

21.4

Peak mem (GB)

Hardware

Chip M5 Pro (20c)

Memory 64 GB

GPU Cores 20

Software

oMLX v0.5.2.dev1

macOS macOS 26.5.2

Context 4,096

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	2,045	73.7	20.5 GB	view
4k	2,457	72.3	21.4 GB	current
8k	2,373	70.3	21.7 GB	view
16k	1,506	65.5	21.5 GB	view
32k	1,015	60.3	21.8 GB	view
64k	626.7	50.4	22.4 GB	view