← Back to community benchmarks

Qwen3.6-35B-A3B

M4 Max (40c) · 64 GB · 4bit · 2026-04-28

Performance

8k

tokens

1,683

PP tok/s

102.3

TG tok/s

4868

TTFT (ms)

20.7

Peak mem (GB)

Hardware

Chip M4 Max (40c)

Memory 64 GB

GPU Cores 40

Software

oMLX v0.3.8rc1

macOS macOS 26.4.1

Context 8,192

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	1,396	359.5	20.6 GB	view
4k	872.7	107.9	20.4 GB	view
8k	1,683	102.3	20.7 GB	current
16k	2,492	99.8	21.9 GB	view
32k	1,733	100.4	24.6 GB	view
64k	1,014	90.7	30.7 GB	view
128k	588.3	80.4	42.9 GB	view