← Back to community benchmarks

Qwen3.6-35B-A3B-oQ4

M4 Max (32c) · 36 GB · 4bit · 2026-04-27

Performance

4k

tokens

515.3

PP tok/s

91.6

TG tok/s

7949

TTFT (ms)

20.7

Peak mem (GB)

Hardware

Chip M4 Max (32c)

Memory 36 GB

GPU Cores 32

Software

oMLX v0.3.6

macOS macOS 26.5

Context 4,096

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	1,162	296.9	20.6 GB	view
4k	515.3	91.6	20.7 GB	current
8k	1,344	85.7	21.1 GB	view
16k	1,254	75.4	21.7 GB	view