← Back to community benchmarks

Qwen3.5-27B

M5 Pro (20c) · 64 GB · 4bit · 2026-03-18
Performance
8k
tokens
446.9
PP tok/s
16.0
TG tok/s
18330
TTFT (ms)
18.9
Peak mem (GB)
Hardware
Chip M5 Pro (20c)
Memory 64 GB
GPU Cores 20
Software
oMLX v0.2.18
macOS macOS 26.3.1
Context 8,192
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 463.2 18.0 16.7 GB view
4k 476.3 17.3 18.3 GB view
8k 446.9 16.0 18.9 GB current
Batching Results
Batch Size TG tok/s Speedup
18.0 1.00×
29.6 1.64×
42.8 2.38×