← Back to community benchmarks

NVIDIA-Nemotron-3-Super-120B-A12B

M4 Max (40c) · 128 GB · 4bit · 2026-04-05
Performance
16k
tokens
395.2
PP tok/s
38.8
TG tok/s
41459
TTFT (ms)
71.8
Peak mem (GB)
Hardware
Chip M4 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.4
macOS macOS 15.7.5
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 371.4 49.4 68.5 GB view
4k 405.7 47.5 71.7 GB view
8k 400.7 43.1 71.7 GB view
16k 395.2 38.8 71.8 GB current
32k 393.0 32.9 71.9 GB view
64k 368.5 22.8 72.1 GB view
128k 317.7 14.1 72.6 GB view