← Back to community benchmarks

Qwen3.6-27B-oQ8-fp16-mtp

M1 Max (32c) · 64 GB · 8bit · 2026-05-06
User commentMTP OFF
Performance
1k
tokens
142.2
PP tok/s
11.3
TG tok/s
7201
TTFT (ms)
28.3
Peak mem (GB)
Hardware
Chip M1 Max (32c)
Memory 64 GB
GPU Cores 32
Software
oMLX v0.3.9.dev1
macOS macOS 26.4.1
Context 1,024
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 142.2 11.3 28.3 GB current
4k 148.3 11.0 29.8 GB view
8k 147.7 10.9 30.8 GB view
16k 144.3 10.5 32.3 GB view
32k 136.1 9.8 35.3 GB view