← Back to community benchmarks

Qwen3.6-35B-A3BoQ5

M2 Max (30c) · 64 GB · 5bit · 2026-04-17

Performance

16k

tokens

784.5

PP tok/s

57.3

TG tok/s

20886

TTFT (ms)

25.6

Peak mem (GB)

Hardware

Chip M2 Max (30c)

Memory 64 GB

GPU Cores 30

Software

oMLX v0.3.6

macOS macOS 26.4.1

Context 16,384

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	817.3	76.1	23.9 GB	view
4k	994.9	71.9	24.6 GB	view
8k	842.9	63.6	25.0 GB	view
16k	784.5	57.3	25.6 GB	current
32k	669.2	44.4	26.9 GB	view