← Back to community benchmarks

Qwen3.6-35B-A3B

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

Performance

32k

tokens

3,118

PP tok/s

56.1

TG tok/s

10508

TTFT (ms)

68.7

Peak mem (GB)

Hardware

Chip M5 Max (40c)

Memory 128 GB

GPU Cores 40

Software

oMLX v0.3.6

macOS macOS 26.5

Context 32,768

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	2,273	154.2	67.0 GB	view
4k	243.4	70.5	66.4 GB	view
8k	3,634	68.0	66.7 GB	view
16k	3,557	64.3	67.4 GB	view
32k	3,118	56.1	68.7 GB	current
64k	2,357	44.0	71.3 GB	view
128k	1,514	33.2	76.8 GB	view