← Back to community benchmarks

Qwen3.6-27B-oQ8-fp16-mtp

M1 Max (32c) · 64 GB · 8bit · 2026-05-06

Performance

4k

tokens

148.2

PP tok/s

14.2

TG tok/s

27636

TTFT (ms)

30.3

Peak mem (GB)

Hardware

Chip M1 Max (32c)

Memory 64 GB

GPU Cores 32

Software

oMLX v0.3.9.dev1

macOS macOS 26.4.1

Context 4,096

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	142.0	14.1	28.8 GB	view
4k	148.2	14.2	30.3 GB	current
8k	147.7	13.4	31.3 GB	view
16k	144.3	13.1	32.8 GB	view