← Back to community benchmarks

Qwen3.6-35B-A3B-oQ8

M5 Max (40c) · 128 GB · 8bit · 2026-04-18

Performance

16k

tokens

3,725

PP tok/s

83.3

TG tok/s

4398

TTFT (ms)

37.1

Peak mem (GB)

Hardware

Chip M5 Max (40c)

Memory 128 GB

GPU Cores 40

Software

oMLX v0.3.6

macOS macOS 26.5

Context 16,384

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	2,149	97.9	35.4 GB	view
4k	3,594	93.5	36.2 GB	view
8k	3,857	89.4	36.5 GB	view
16k	3,725	83.3	37.1 GB	current
32k	3,194	74.2	38.5 GB	view
64k	2,360	57.5	41.2 GB	view
128k	1,520	38.2	46.5 GB	view