← Back to community benchmarks

Youssofal_Qwen3.6-35B-A3B-Abliterated-Heretic

M5 Max (40c) · 128 GB · 4bit · 2026-07-25

Performance

64k

tokens

2,363

PP tok/s

83.5

TG tok/s

27735

TTFT (ms)

29.0

Peak mem (GB)

Hardware

Chip M5 Max (40c)

Memory 128 GB

GPU Cores 40

Software

oMLX v0.5.3

macOS macOS 26.5.1

Context 65,536

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	3,687	108.7	23.3 GB	view
4k	4,232	108.8	24.1 GB	view
64k	2,363	83.5	29.0 GB	current

Batching Results

Batch Size	TG tok/s	Speedup
1×	108.7	1.00×
2×	203.6	1.87×
4×	419.4	3.86×