Can a Jetson Orin NX 8GB run Gemma 2 9B?

Yes, usable but slower than reading speed. Measured generation speed: 8.8 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.

BoardSizeQuantRuntimeGen tok/sPrompt tok/sTTFT msSource
Jetson Orin NX 8GB
Jetson Orin NX 8GB (original mode)
9BINT4MLC8.8––source ↗
2025-01-16

Gemma 2 9B on other boards

Buy a Jetson Orin NX 8GB

Ran Gemma 2 9B on a Jetson Orin NX 8GB? Submit your numbers →

Stay in the loop.

New boards, new runtimes, new numbers. One email a month, no hype.