← Back to community benchmarks

Qwen3.6-27B-oQ8-mtp

M3 Ultra (80c) · 96 GB · 8bit · 2026-05-12

Performance

4k

tokens

422.8

PP tok/s

34.5

TG tok/s

9688

TTFT (ms)

30.3

Peak mem (GB)

Hardware

Chip M3 Ultra (80c)

Memory 96 GB

GPU Cores 80

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	389.1	34.4	28.8 GB	view
4k	422.8	34.5	30.3 GB	current

Batching Results

Batch Size	TG tok/s	Speedup
1×	34.4	1.00×
2×	42.1	1.22×
4×	61.6	1.79×