← Back to community benchmarks

Qwen3.6-35B-A3B-mxfp8

M3 Max (40c) · 128 GB · 8bit · 2026-04-17

Performance

4k

tokens

1,506

PP tok/s

70.9

TG tok/s

2719

TTFT (ms)

35.2

Peak mem (GB)

Hardware

Chip M3 Max (40c)

Memory 128 GB

GPU Cores 40

Software

oMLX v0.3.6

macOS macOS 26.3.1

Context 4,096

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
4k	1,506	70.9	35.2 GB	current
8k	1,501	68.0	35.5 GB	view
16k	1,399	61.8	36.1 GB	view