← Back to community benchmarks

Qwen3.6-27B-UD

M5 Max (40c) · 128 GB · 6bit · 2026-05-05
Performance
128k
tokens
419.1
PP tok/s
13.2
TG tok/s
312752
TTFT (ms)
54.7
Peak mem (GB)
Hardware
Chip M5 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.8
macOS macOS 26.4.1
Context 131,072
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 507.1 18.3 29.2 GB view
4k 675.7 18.3 30.6 GB view
64k 568.2 15.0 40.1 GB view
128k 419.1 13.2 54.7 GB current
Batching Results
Batch Size TG tok/s Speedup
18.3 1.00×
21.2 1.16×
34.0 1.86×
45.4 2.48×