← Back to community benchmarks

Qwen3.6-27B-oQ8-mtp

M5 Max (40c) · 128 GB · 8bit · 2026-05-09
Performance
16k
tokens
643.3
PP tok/s
13.3
TG tok/s
25471
TTFT (ms)
31.7
Peak mem (GB)
Hardware
Chip M5 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.6
macOS macOS 26.4.1
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 720.1 18.2 28.3 GB view
4k 813.6 13.9 29.8 GB view
8k 705.2 17.3 30.4 GB view
16k 643.3 13.3 31.7 GB current
32k 568.2 14.2 34.2 GB view
Batching Results
Batch Size TG tok/s Speedup
18.2 1.00×
32.2 1.77×
36.8 2.02×