← Back to community benchmarks

Qwen3.6-27B-oQ4-fp16-mtp

M1 Ultra (48c) · 128 GB · 4bit · 2026-05-18

Performance

4k

tokens

226.7

PP tok/s

16.7

TG tok/s

18064

TTFT (ms)

17.9

Peak mem (GB)

Hardware

Chip M1 Ultra (48c)

Memory 128 GB

GPU Cores 48

Software

oMLX v0.3.8

macOS macOS 26.4.1

Context 4,096

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	212.9	28.7	16.4 GB	view
4k	226.7	16.7	17.9 GB	current

Batching Results

Batch Size	TG tok/s	Speedup
1×	28.7	1.00×
2×	32.5	1.13×
4×	40.1	1.40×