← Back to community benchmarks

Qwen3.5-27B

M2 Ultra (76c) · 128 GB · 4bit · 2026-03-08
Performance
16k
tokens
222.3
PP tok/s
26.6
TG tok/s
73711
TTFT (ms)
18.3
Peak mem (GB)
Hardware
Chip M2 Ultra (76c)
Memory 128 GB
GPU Cores 76
Software
oMLX v0.2.6
macOS macOS 26.3
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 226.5 33.4 16.6 GB view
4k 228.9 31.3 17.0 GB view
8k 227.0 29.3 17.4 GB view
16k 222.3 26.6 18.3 GB current
32k 212.9 21.9 20.1 GB view
64k 192.2 16.8 23.6 GB view
Batching Results
Batch Size TG tok/s Speedup
33.4 1.00×
47.7 1.43×
53.5 1.60×
58.9 1.76×