← Back to community benchmarks

Qwen3.6-27B-OptiQ

M1 Max (24c) · 32 GB · 4bit · 2026-06-04
Performance
8k
tokens
66.3
PP tok/s
11.5
TG tok/s
123536
TTFT (ms)
21.7
Peak mem (GB)
Hardware
Chip M1 Max (24c)
Memory 32 GB
GPU Cores 24
Software
oMLX v0.4.2.dev2
macOS macOS 26.5.1
Context 8,192
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 62.7 12.7 18.6 GB view
4k 65.3 10.1 19.5 GB view
8k 66.3 11.5 21.7 GB current
16k 58.3 10.9 22.5 GB view
32k 58.9 10.0 22.5 GB view
Batching Results
Batch Size TG tok/s Speedup
12.7 1.00×
14.8 1.17×
16.9 1.33×