← Back to community benchmarks

Qwen3.6-27B-MTP

M5 Max (40c) · 128 GB · 8bit · 2026-05-17
Performance
1k
tokens
702.1
PP tok/s
28.7
TG tok/s
1459
TTFT (ms)
28.8
Peak mem (GB)
Hardware
Chip M5 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.9.dev2
macOS macOS 26.5
Context 1,024
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 702.1 28.7 28.8 GB current
4k 789.6 28.6 30.2 GB view
8k 753.6 27.7 31.2 GB view
Batching Results
Batch Size TG tok/s Speedup
28.7 1.00×
33.3 1.16×
47.3 1.65×