← Back to community benchmarks

Qwen3.6-27B-oQ8-mtp

M3 Ultra (80c) · 512 GB · 8bit · 2026-05-13

Performance

64k

tokens

333.3

PP tok/s

23.7

TG tok/s

196625

TTFT (ms)

39.7

Peak mem (GB)

Hardware

Chip M3 Ultra (80c)

Memory 512 GB

GPU Cores 80

Software

oMLX v0.3.9.dev2

macOS macOS 26.4.1

Context 65,536

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
64k	333.3	23.7	39.7 GB	current
128k	253.1	19.5	54.2 GB	view
195k	201.0	16.5	68.9 GB	view