Jetson Orin NX 16GB Super / Phi-3.5 3.8B
Can a Jetson Orin NX 16GB Super run Phi-3.5 3.8B?
Yes, at a comfortable chat speed. Measured generation speed: 40.9 tok/s across 1 public run.
Rule of thumb: 10+ tok/s feels like a chat; 4–10 is usable for assistants and automations; under 4 is batch-job territory. Speeds depend on quantization, runtime, context length and cooling. Check each source.
| Board | Quant | Runtime | Gen tok/s | Prompt tok/s | TTFT ms | Source |
|---|---|---|---|---|---|---|
| Jetson Orin NX 16GB Super Jetson Orin NX 16GB (Super mode) | INT4 | MLC | 40.9 | – | – | source ↗ 2025-01-16 |
Phi-3.5 3.8B on other boards
- Jetson Orin NX 8GB Super 41.3 tok/s best
- Jetson Orin Nano Super 38.1 tok/s best
- Jetson Orin NX 16GB 35.9 tok/s best
- Jetson Orin NX 8GB 30.8 tok/s best
- Jetson Orin Nano 24.7 tok/s best
Buy a Jetson Orin NX 16GB Super
- Jetson Orin NX 16GB module (Super mode): price not verified yet Check price (affiliate slot)
Ran Phi-3.5 3.8B on a Jetson Orin NX 16GB Super? Submit your numbers →