← Back to community benchmarks

Qwen3.5-27B

M5 Max (40c) · 128 GB · 8bit · 2026-04-12

Performance

64k

tokens

559.7

PP tok/s

13.8

TG tok/s

117099

TTFT (ms)

39.2

Peak mem (GB)

Hardware

Chip M5 Max (40c)

Memory 128 GB

GPU Cores 40

Software

oMLX v0.3.5.dev2

macOS macOS 26.4.1

Context 65,536

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
16k	695.0	16.0	31.7 GB	view
32k	658.1	15.5	34.2 GB	view
64k	559.7	13.8	39.2 GB	current