← Back to community benchmarks

mtp-qwen36-27b

M2 Ultra (60c) · 128 GB · 4bit · 2026-05-10

Performance

1k

tokens

174.1

PP tok/s

30.5

TG tok/s

5882

TTFT (ms)

16.8

Peak mem (GB)

Hardware

Chip M2 Ultra (60c)

Memory 128 GB

GPU Cores 60

Software

oMLX v0.3.9.dev1

macOS macOS 26.4.1

Context 1,024

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	174.1	30.5	16.8 GB	current
4k	181.7	31.3	18.2 GB	view

Batching Results

Batch Size	TG tok/s	Speedup
1×	30.5	1.00×
2×	38.6	1.27×
4×	43.6	1.43×