← Back to community benchmarks

Qwen3.5-27B

M2 Pro (16c) · 32 GB · 4bit · 2026-03-11
Performance
1k
tokens
48.5
PP tok/s
11.1
TG tok/s
21103
TTFT (ms)
16.6
Peak mem (GB)
Hardware
Chip M2 Pro (16c)
Memory 32 GB
GPU Cores 16
Software
oMLX v0.2.7
macOS macOS 15.6.1
Context 1,024
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 48.5 11.1 16.6 GB current
4k 48.8 10.2 17.0 GB view
Batching Results
Batch Size TG tok/s Speedup
11.1 1.00×
12.9 1.16×
13.2 1.19×