← Back to community benchmarks

Qwen3.5-27B

M4 (10c) · 32 GB · 4bit · 2026-03-11
Performance
4k
tokens
61.3
PP tok/s
6.4
TG tok/s
66812
TTFT (ms)
16.9
Peak mem (GB)
Hardware
Chip M4 (10c)
Memory 32 GB
GPU Cores 10
Software
oMLX v0.2.7
macOS macOS 15.7.4
Context 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 60.6 6.4 16.6 GB view
4k 61.3 6.4 16.9 GB current
8k 58.7 6.2 17.3 GB view
16k 56.7 5.7 18.2 GB view
Batching Results
Batch Size TG tok/s Speedup
6.4 1.00×
13.3 2.08×
16.3 2.55×
16.5 2.58×