← Back to community benchmarks

NVIDIA-Nemotron-3-Nano-30B-A3B

M3 Max (40c) · 48 GB · 4bit · 2026-03-11
Performance
4k
tokens
1,307
PP tok/s
119.5
TG tok/s
3135
TTFT (ms)
18.3
Peak mem (GB)
Hardware
Chip M3 Max (40c)
Memory 48 GB
GPU Cores 40
Software
oMLX v0.2.8
macOS macOS 26.3
Context 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 974.1 125.4 18.2 GB view
4k 1,307 119.5 18.3 GB current
Batching Results
Batch Size TG tok/s Speedup
125.4 1.00×
183.6 1.46×
234.9 1.87×