← Back to community benchmarks

Qwen3.6-35B-A3B8bit

M2 Max (38c) · 96 GB · 8bit · 2026-05-04

Performance

4k

tokens

1,173

PP tok/s

174.9

TG tok/s

3491

TTFT (ms)

40.5

Peak mem (GB)

Hardware

Chip M2 Max (38c)

Memory 96 GB

GPU Cores 38

Software

oMLX v0.3.8

macOS macOS 26.4.1

Context 4,096

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	1,181	190.4	36.8 GB	view
4k	1,173	174.9	40.5 GB	current
8k	1,156	146.6	41.4 GB	view
16k	1,089	109.9	43.1 GB	view
32k	939.0	71.8	46.3 GB	view
64k	715.8	41.3	52.8 GB	view
128k	472.2	23.7	65.7 GB	view
195k	311.5	15.3	78.9 GB	view