← Back to community benchmarks

Nemotron-3-Super-120B-A12B3.6bit

M3 Ultra (60c) · 96 GB · 3bit · 2026-04-07

Performance

16k

tokens

1,993

PP tok/s

57.6

TG tok/s

8221

TTFT (ms)

59.7

Peak mem (GB)

Hardware

Chip M3 Ultra (60c)

Memory 96 GB

GPU Cores 60

Software

oMLX v0.3.4

macOS macOS 26.4

Context 16,384

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
4k	540.1	56.7	60.0 GB	view
16k	1,993	57.6	59.7 GB	current
64k	1,750	54.4	59.8 GB	view