← Back to community benchmarks

Qwen3.5-27B

M3 Max (30c) · 96 GB · 4bit · 2026-04-15

Performance

1k

tokens

147.1

PP tok/s

58.9

TG tok/s

6961

TTFT (ms)

31.3

Peak mem (GB)

Hardware

Chip M3 Max (30c)

Memory 96 GB

GPU Cores 30

Software

oMLX v0.3.5

macOS macOS 26.4.1

Context 1,024