← Back to community benchmarks

Qwen3.6-35B-A3B

M3 Ultra (60c) · 96 GB · 4bit · 2026-04-24

Performance

8k

tokens

2,155

PP tok/s

94.3

TG tok/s

3801

TTFT (ms)

20.9

Peak mem (GB)

Hardware

Chip M3 Ultra (60c)

Memory 96 GB

GPU Cores 60

Software

oMLX v0.3.7rc2

macOS macOS 26.4.1

Context 8,192

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
8k	2,155	94.3	20.9 GB	current
16k	4,155	52.0	22.1 GB	view
32k	3,913	50.8	22.7 GB	view
64k	3,219	48.9	24.6 GB	view