← Back to community benchmarks

Qwen3.6-35B-A3B-YOYO-V2-oQ4e-mtp

M5 Max (40c) · 128 GB · 4bit · 2026-07-17

Performance

64k

tokens

2,354

PP tok/s

83.8

TG tok/s

27836

TTFT (ms)

25.8

Peak mem (GB)

Hardware

Chip M5 Max (40c)

Memory 128 GB

GPU Cores 40

Software

oMLX v0.5.1

macOS macOS 26.5.2

Context 65,536

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	3,400	115.7	19.9 GB	view
4k	4,005	114.8	20.7 GB	view
64k	2,354	83.8	25.8 GB	current

Batching Results

Batch Size	TG tok/s	Speedup
1×	115.7	1.00×
2×	187.7	1.62×
4×	290.9	2.51×