← Back to community benchmarks

Qwen3.6-35B-A3B-oQ4-mtp

M3 Ultra (80c) · 512 GB · 4bit · 2026-06-08

Performance

32k

tokens

2,065

PP tok/s

85.0

TG tok/s

15871

TTFT (ms)

23.5

Peak mem (GB)

Hardware

Chip M3 Ultra (80c)

Memory 512 GB

GPU Cores 80

Software

oMLX v0.4.2rc1

macOS macOS 26.4

Context 32,768

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	2,179	95.7	20.4 GB	view
4k	2,783	99.4	21.2 GB	view
8k	2,543	87.7	21.5 GB	view
16k	2,403	86.6	22.1 GB	view
32k	2,065	85.0	23.5 GB	current
64k	1,559	65.9	26.2 GB	view