← Back to community benchmarks

Qwen3.6-27B-MTPLX-Optimized-Speed

M1 Max (24c) · 64 GB · 4bit · 2026-05-20
Performance
8k
tokens
30.0
PP tok/s
6.8
TG tok/s
273463
TTFT (ms)
18.3
Peak mem (GB)
Hardware
Chip M1 Max (24c)
Memory 64 GB
GPU Cores 24
Software
oMLX v0.3.8
macOS macOS 26.4
Context 8,192
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 42.4 7.3 15.8 GB view
4k 30.7 7.0 17.3 GB view
8k 30.0 6.8 18.3 GB current
Batching Results
Batch Size TG tok/s Speedup
7.3 1.00×
7.6 1.04×
8.4 1.15×