← Back to community benchmarks

qw3.6-27b-ud

M5 Max (40c) · 64 GB · 4bit · 2026-05-01

Performance

1k

tokens

787.4

PP tok/s

21.4

TG tok/s

1300

TTFT (ms)

25.2

Peak mem (GB)

Hardware

Chip M5 Max (40c)

Memory 64 GB

GPU Cores 40

Software

oMLX v0.3.8

macOS macOS 26.4.1

Context 1,024

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	787.4	21.4	25.2 GB	current
4k	785.7	21.1	26.6 GB	view

Batching Results

Batch Size	TG tok/s	Speedup
1×	21.4	1.00×
2×	24.4	1.14×
4×	39.9	1.86×