← Back to community benchmarks

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

M2 Ultra (60c) · 64 GB · 4bit · 2026-05-20
Performance
4k
tokens
296.1
PP tok/s
36.4
TG tok/s
13832
TTFT (ms)
18.2
Peak mem (GB)
Hardware
Chip M2 Ultra (60c)
Memory 64 GB
GPU Cores 60
Software
oMLX v0.3.9rc1
macOS macOS 26.3.1
Context 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 276.0 34.8 16.7 GB view
4k 296.1 36.4 18.2 GB current
8k 295.4 32.8 19.2 GB view
16k 288.2 33.1 20.7 GB view
32k 272.9 29.4 23.7 GB view
64k 229.9 24.6 29.7 GB view
Batching Results
Batch Size TG tok/s Speedup
34.8 1.00×
46.0 1.32×
53.5 1.54×
59.3 1.70×