Granite-Vision-4.1-4B on Jetson Orin NX 16GB — Q8_0, llama.cpp vision-language inference benchmark

Open this configuration in the VLM Inference Benchmark Explorer

Model	Granite-Vision-4.1-4B
Parameters	4.0B
Device	Jetson Orin NX 16GB
Quantization	Q8_0
Image size	720p
Images per camera	1
Max Cameras (response ≤ 3 s)	>3 s
Engine	llama.cpp

Cameras	Response (s)	TPS (tok/s)	Target
1	4.36	14.4	>3 s
2	4.78	11.4	>3 s
4	19.0	11.2	>3 s
8	44.1	12.0	>3 s

Images per request	Response (s)	TPS (tok/s)	Target
1	4.36	14.4	>3 s
2	8.76	13.9	>3 s
4	17.9	13.2	>3 s
8	37.9	11.6	>3 s
16	85.0	8.8	>3 s
32	—	—	Failed

Granite-Vision-4.1-4B (4.0B parameters), served in Q8_0 format with llama.cpp on Jetson Orin NX 16GB, starts answering a single camera that sends one 720p image after 4.36 s, then writes 14.4 tok/s.

With 16 images per camera at 720p, the answer starts after 84.98 s. With one 2K image, after 4.52 s.

At the default target of an answer starting within 3 s (720p, one image per camera), not even one camera is answered in time.