← Back to community benchmarks

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

M1 Max (32c) · 32 GB · 4bit · 2026-05-14
Performance
4k
tokens
151.6
PP tok/s
21.9
TG tok/s
27026
TTFT (ms)
18.2
Peak mem (GB)
Hardware
Chip M1 Max (32c)
Memory 32 GB
GPU Cores 32
Software
oMLX v0.3.9.dev2
macOS macOS 26.4.1
Context 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 125.7 21.1 16.7 GB view
4k 151.6 21.9 18.2 GB current
Batching Results
Batch Size TG tok/s Speedup
1× 21.1 1.00×
2× 27.9 1.32×
4× 30.2 1.43×