← Back to community benchmarks

Qwen3.6-35B-A3B-oQ4

M4 Max (40c) · 48 GB · 4bit · 2026-04-16

Performance

8k

tokens

1,736

PP tok/s

107.3

TG tok/s

4718

TTFT (ms)

21.1

Peak mem (GB)

Hardware

Chip M4 Max (40c)

Memory 48 GB

GPU Cores 40

Software

oMLX v0.3.6

macOS macOS 26.5

Context 8,192

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	1,397	114.2	19.9 GB	view
4k	1,743	111.2	20.7 GB	view
8k	1,736	107.3	21.1 GB	current
16k	1,578	97.3	21.7 GB	view