← Back to community benchmarks

Qwen3.6-35B-A3B-Hereticmixed-6.4bit

M4 Pro (16c) · 48 GB · 6bit · 2026-04-30

Performance

32k

tokens

574.0

PP tok/s

38.4

TG tok/s

57091

TTFT (ms)

29.7

Peak mem (GB)

Hardware

Chip M4 Pro (16c)

Memory 48 GB

GPU Cores 16

Software

oMLX v0.3.8rc1

macOS macOS 26.4.1

Context 32,768

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	640.9	59.3	26.6 GB	view
4k	727.4	56.2	27.4 GB	view
8k	713.9	53.3	27.7 GB	view
16k	663.4	47.1	28.3 GB	view
32k	574.0	38.4	29.7 GB	current
64k	446.3	28.0	32.4 GB	view