← Back to community benchmarks

Qwen3.5-27BoQ4

M3 Max (40c) · 128 GB · 4bit · 2026-04-18
Performance
16k
tokens
648.7
PP tok/s
15.1
TG tok/s
25255
TTFT (ms)
20.8
Peak mem (GB)
Hardware
Chip M3 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.6
macOS macOS 26.4.1
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 197.1 81.3 20.1 GB view
4k 170.8 18.9 18.4 GB view
8k 168.2 18.5 19.1 GB view
16k 648.7 15.1 20.8 GB current
32k 628.5 14.7 22.1 GB view