← Back to community benchmarks

Qwen3.6-27B-oQ6-mtp

M4 Pro (20c) · 64 GB · 6bit · 2026-05-17
Performance
4k
tokens
78.1
PP tok/s
8.4
TG tok/s
52455
TTFT (ms)
24.4
Peak mem (GB)
Hardware
Chip M4 Pro (20c)
Memory 64 GB
GPU Cores 20
Software
oMLX v0.3.9.dev2
macOS macOS 26.4.1
Context 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 80.5 8.7 23.0 GB view
4k 78.1 8.4 24.4 GB current
Batching Results
Batch Size TG tok/s Speedup
8.7 1.00×
9.3 1.07×
11.4 1.31×