← Back to community benchmarks

Qwen3.6-27B-Uncensored-Heretic-v2

M5 Max (40c) · 64 GB · 6bit · 2026-06-19
Performance
128k
tokens
421.9
PP tok/s
16.5
TG tok/s
310702
TTFT (ms)
40.7
Peak mem (GB)
Hardware
Chip M5 Max (40c)
Memory 64 GB
GPU Cores 40
Software
oMLX v0.4.4
macOS macOS 26.5.1
Context 131,072
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 715.9 24.5 22.1 GB view
4k 852.6 23.7 23.5 GB view
8k 775.5 24.1 24.2 GB view
16k 700.2 23.4 25.4 GB view
32k 641.6 22.0 27.9 GB view
64k 557.7 19.4 32.9 GB view
128k 421.9 16.5 40.7 GB current
Batching Results
Batch Size TG tok/s Speedup
24.5 1.00×
36.9 1.51×
92.9 3.79×