← Back to community benchmarks

Qwen3.6-27B-UD

M5 Max (40c) · 128 GB · 4bit · 2026-04-29
Performance
128k
tokens
390.1
PP tok/s
11.9
TG tok/s
336012
TTFT (ms)
46.5
Peak mem (GB)
Hardware
Chip M5 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.7rc1
macOS macOS 26.4.1
Context 131,072
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 762.9 21.3 25.2 GB view
4k 899.6 20.8 26.6 GB view
8k 875.0 20.4 27.3 GB view
16k 739.1 19.5 28.5 GB view
32k 681.5 18.2 31.0 GB view
64k 534.5 15.2 36.1 GB view
128k 390.1 11.9 46.5 GB current
195k 298.2 8.3 57.2 GB view
Batching Results
Batch Size TG tok/s Speedup
21.3 1.00×
24.6 1.15×