← Back to community benchmarks

Qwen3.6-35B-A3B-oQ4

M3 Max (40c) · 128 GB · 4bit · 2026-04-26

Performance

4k

tokens

720.3

PP tok/s

84.4

TG tok/s

5687

TTFT (ms)

21.1

Peak mem (GB)

Hardware

Chip M3 Max (40c)

Memory 128 GB

GPU Cores 40

Software

oMLX v0.3.7rc2

macOS macOS 26.4.1

Context 4,096

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	1,205	153.3	21.3 GB	view
4k	720.3	84.4	21.1 GB	current
16k	2,776	33.4	22.6 GB	view