← Back to community benchmarks

Qwen3.6-35B-A3B-oQ6

M4 Max (40c) · 64 GB · 6bit · 2026-04-19

Performance

32k

tokens

1,200

PP tok/s

51.7

TG tok/s

27299

TTFT (ms)

30.8

Peak mem (GB)

Hardware

Chip M4 Max (40c)

Memory 64 GB

GPU Cores 40

Software

oMLX v0.3.6

macOS macOS 26.3.1

Context 32,768

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	1,372	285.5	29.1 GB	view
4k	549.9	88.3	28.5 GB	view
8k	1,663	84.2	28.9 GB	view
16k	1,523	76.8	29.5 GB	view
32k	1,200	51.7	30.8 GB	current
64k	983.1	33.4	33.5 GB	view
128k	542.0	19.2	38.9 GB	view