← Back to community benchmarks

Qwen3.6-27B-oQ4-fp16-mtp

M3 Max (40c) · 48 GB · 4bit · 2026-05-16
Performance
1k
tokens
221.7
PP tok/s
32.8
TG tok/s
4619
TTFT (ms)
16.7
Peak mem (GB)
Hardware
Chip M3 Max (40c)
Memory 48 GB
GPU Cores 40
Software
oMLX v0.3.9.dev1
macOS macOS 26.5
Context 1,024
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 221.7 32.8 16.7 GB current
4k 224.6 30.1 18.2 GB view
Batching Results
Batch Size TG tok/s Speedup
32.8 1.00×
39.1 1.19×
47.3 1.44×