← Back to community benchmarks

Nex-N2-Pro-Q5.3

M3 Ultra (80c) · 512 GB · 4bit · 2026-07-17

Performance

128k

tokens

291.3

PP tok/s

20.3

TG tok/s

450010

TTFT (ms)

256.4

Peak mem (GB)

Hardware

Chip M3 Ultra (80c)

Memory 512 GB

GPU Cores 80

Software

oMLX v0.5.1

macOS macOS 26.5.2

Context 131,072

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	483.7	29.8	248.7 GB	view
4k	551.6	29.3	250.3 GB	view
8k	543.0	28.7	250.9 GB	view
16k	516.0	27.6	251.5 GB	view
32k	464.2	26.3	253.2 GB	view
64k	386.9	23.8	253.8 GB	view
128k	291.3	20.3	256.4 GB	current