← Back to community benchmarks

Qwen3.6-35B-A3B

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

Performance

32k

tokens

3,917

PP tok/s

50.8

TG tok/s

8366

TTFT (ms)

22.7

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 32,768

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
8k	2,157	94.6	20.9 GB	view
16k	4,155	52.2	22.1 GB	view
32k	3,917	50.8	22.7 GB	current
64k	3,220	49.0	24.6 GB	view