← Back to community benchmarks

Qwen3.6-27B

M1 Ultra (48c) · 64 GB · 4bit · 2026-06-04

Performance

1k

tokens

134.9

PP tok/s

22.2

TG tok/s

7592

TTFT (ms)

15.1

Peak mem (GB)

Hardware

Chip M1 Ultra (48c)

Memory 64 GB

GPU Cores 48

Software

oMLX v0.4.0rc2

macOS macOS 26.5

Context 1,024

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	134.9	22.2	15.1 GB	current
4k	140.2	21.3	16.5 GB	view

Batching Results

Batch Size	TG tok/s	Speedup
1×	22.2	1.00×
2×	26.4	1.19×
4×	29.2	1.32×