← Back to community benchmarks

Qwen3.5-27B

M4 Max (40c) · 64 GB · 4bit · 2026-03-23

Performance

4k

tokens

219.4

PP tok/s

25.6

TG tok/s

18671

TTFT (ms)

18.3

Peak mem (GB)

Hardware

Chip M4 Max (40c)

Memory 64 GB

GPU Cores 40

Software

oMLX v0.2.20rc1

macOS macOS 15.7.4

Context 4,096

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	250.0	28.6	16.7 GB	view
4k	219.4	25.6	18.3 GB	current

Batching Results

Batch Size	TG tok/s	Speedup
1×	28.6	1.00×
2×	43.3	1.51×
4×	48.2	1.69×