← Back to community benchmarks

Qwen3.6-27B

M5 (10c) · 32 GB · 4bit · 2026-05-20
Performance
16k
tokens
128.7
PP tok/s
6.8
TG tok/s
127340
TTFT (ms)
18.3
Peak mem (GB)
Hardware
Chip M5 (10c)
Memory 32 GB
GPU Cores 10
Software
oMLX v0.3.6
macOS macOS 26.4
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 192.4 8.8 15.0 GB view
4k 192.4 8.5 16.4 GB view
8k 173.8 8.2 17.0 GB view
16k 128.7 6.8 18.3 GB current
Batching Results
Batch Size TG tok/s Speedup
8.8 1.00×
13.8 1.57×
27.8 3.16×