← Back to community benchmarks

Qwen3.6-27B-oQ5-mtp

M3 Ultra (60c) · 256 GB · 5bit · 2026-05-11

Performance

4k

tokens

311.7

PP tok/s

33.0

TG tok/s

13140

TTFT (ms)

21.4

Peak mem (GB)

Hardware

Chip M3 Ultra (60c)

Memory 256 GB

GPU Cores 60

Software

oMLX v0.3.9.dev1

macOS macOS 15.6.1

Context 4,096

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	295.5	33.5	19.9 GB	view
4k	311.7	33.0	21.4 GB	current

Batching Results

Batch Size	TG tok/s	Speedup
1×	33.5	1.00×
2×	41.4	1.24×
4×	46.2	1.38×