← Back to community benchmarks

Qwen3.6-27B

M5 Max (40c) · 64 GB · 8bit · 2026-06-04
Performance
4k
tokens
815.8
PP tok/s
18.0
TG tok/s
5021
TTFT (ms)
29.7
Peak mem (GB)
Hardware
Chip M5 Max (40c)
Memory 64 GB
GPU Cores 40
Software
oMLX v0.4.1
macOS macOS 26.3.2
Context 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 723.5 18.1 28.3 GB view
4k 815.8 18.0 29.7 GB current
8k 821.5 17.5 30.8 GB view
16k 721.8 16.0 32.3 GB view
Batching Results
Batch Size TG tok/s Speedup
18.1 1.00×
30.9 1.71×
46.2 2.55×