← Back to community benchmarks

Qwen3.6-27Bmxfp8

M5 Max (40c) · 128 GB · 8bit · 2026-04-23
Performance
16k
tokens
749.3
PP tok/s
17.1
TG tok/s
21867
TTFT (ms)
98.3
Peak mem (GB)
Hardware
Chip M5 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.7.dev1
macOS macOS 26.5
Context 16,384