← Back to community benchmarks

Qwen3.6-27BoQ4

M3 Ultra (60c) · 96 GB · 4bit · 2026-04-25
Performance
1k
tokens
322.8
PP tok/s
38.1
TG tok/s
3173
TTFT (ms)
16.6
Peak mem (GB)
Hardware
Chip M3 Ultra (60c)
Memory 96 GB
GPU Cores 60
Software
oMLX v0.3.6
macOS macOS 26.4.1
Context 1,024
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 322.8 38.1 16.6 GB current
8k 344.1 34.7 18.7 GB view
32k 313.3 24.2 22.4 GB view
64k 272.8 16.1 27.5 GB view
Batching Results
Batch Size TG tok/s Speedup
38.1 1.00×
61.8 1.62×
70.8 1.86×
79.3 2.08×