← Back to community benchmarks

Qwen3.6-35B-A3B-Abliterated-Heretic

M4 Max (40c) · 48 GB · 4bit · 2026-04-20

Performance

4k

tokens

1,675

PP tok/s

99.1

TG tok/s

2446

TTFT (ms)

26.2

Peak mem (GB)

Hardware

Chip M4 Max (40c)

Memory 48 GB

GPU Cores 40

Software

oMLX v0.3.6

macOS macOS 26.4.1

Context 4,096

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	1,226	102.3	25.4 GB	view
4k	1,675	99.1	26.2 GB	current
8k	1,679	92.2	26.5 GB	view
16k	1,150	45.0	27.6 GB	view
32k	1,032	44.1	28.8 GB	view