← Back to community benchmarks

Qwen3.6-27B-oQ4-mtp

M3 Max (30c) · 36 GB · 4bit · 2026-05-18
Performance
8k
tokens
136.6
PP tok/s
17.7
TG tok/s
59989
TTFT (ms)
19.2
Peak mem (GB)
Hardware
Chip M3 Max (30c)
Memory 36 GB
GPU Cores 30
Software
oMLX v0.3.9.dev2
macOS macOS 26.5
Context 8,192
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 118.6 14.3 16.7 GB view
4k 126.7 16.7 18.2 GB view
8k 136.6 17.7 19.2 GB current
16k 135.0 16.4 20.7 GB view
Batching Results
Batch Size TG tok/s Speedup
14.3 1.00×
18.2 1.27×