← Back to community benchmarks

Qwen3.6-27B-OptiQ

M4 (10c) · 32 GB · 4bit · 2026-04-27
Performance
1k
tokens
55.0
PP tok/s
5.8
TG tok/s
18618
TTFT (ms)
17.0
Peak mem (GB)
Hardware
Chip M4 (10c)
Memory 32 GB
GPU Cores 10
Software
oMLX v0.3.6
macOS macOS 15.3
Context 1,024
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 55.0 5.8 17.0 GB current
4k 53.5 5.9 18.4 GB view
Batching Results
Batch Size TG tok/s Speedup
5.8 1.00×
10.8 1.86×
19.0 3.28×