← Back to community benchmarks

Qwen3.6-27B-oQ8-mtp

M3 Max (40c) · 128 GB · 8bit · 2026-05-08
Performance
4k
tokens
168.7
PP tok/s
15.2
TG tok/s
24284
TTFT (ms)
30.3
Peak mem (GB)
Hardware
Chip M3 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.9.dev1
macOS macOS 26.3.1
Context 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 163.3 15.3 28.8 GB view
4k 168.7 15.2 30.3 GB current
Batching Results
Batch Size TG tok/s Speedup
1× 15.3 1.00×
2× 17.3 1.13×
4× 23.2 1.52×