Jetson Orin NX 16GB / Llama 3.1 8B
Can a Jetson Orin NX 16GB run Llama 3.1 8B?
Yes, at a comfortable chat speed. Measured generation speed: 20.5 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 Jetson Orin NX 16GB (original mode) | INT4 | MLC | 20.5 | – | – | source ↗ 2025-01-16 |
Llama 3.1 8B on other boards
- Jetson Orin NX 8GB Super 23.1 tok/s best
- Jetson Orin NX 16GB Super 22.8 tok/s best
- Jetson Orin Nano Super 19.1 tok/s best
- Jetson Orin NX 8GB 15.9 tok/s best
- Jetson Orin Nano 14 tok/s best
- Rockchip RK3588 3.72 tok/s best
Buy a Jetson Orin NX 16GB
- Jetson Orin NX 16GB module: price not verified yet Check price (affiliate slot)
Ran Llama 3.1 8B on a Jetson Orin NX 16GB? Submit your numbers →