← Back to community benchmarks
NVIDIA-Nemotron-3-Nano-30B-A3B
Performance
8k
tokens
1,475
PP tok/s
131.8
TG tok/s
5554
TTFT (ms)
18.3
Peak mem (GB)
Hardware
Chip
M4 Max (40c)
Memory
64 GB
GPU Cores
40
Software
oMLX
v0.2.7
macOS
macOS 26.3.1
Context
8,192
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,200 | 141.8 | 18.2 GB | view |
| 4k | 1,519 | 136.4 | 18.3 GB | view |
| 8k | 1,475 | 131.8 | 18.3 GB | current |
| 16k | 1,413 | 114.7 | 18.3 GB | view |
| 32k | 1,135 | 99.3 | 18.4 GB | view |
| 64k | 956.1 | 80.0 | 18.6 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 141.8 | 1.00× |
| 2× | 205.4 | 1.45× |
| 4× | 257.9 | 1.82× |
| 8× | 293.2 | 2.07× |