← Back to community benchmarks

Qwen3.5-27B

M4 (10c) · 32 GB · 4bit · 2026-03-11
Performance
1k
tokens
59.8
PP tok/s
6.1
TG tok/s
17134
TTFT (ms)
16.6
Peak mem (GB)
Hardware
Chip M4 (10c)
Memory 32 GB
GPU Cores 10
Software
oMLX v0.2.7
macOS macOS 15.3.1
Context 1,024
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 59.8 6.1 16.6 GB current
4k 59.7 6.3 16.9 GB view
Batching Results
Batch Size TG tok/s Speedup
6.1 1.00×
12.9 2.11×
15.5 2.54×