Qwen3.5-27B on Jetson AGX Orin 32GB Developer Kit — Q4_K_M, llama.cpp vision-language inference benchmark

Open this configuration in the VLM Inference Benchmark Explorer

Model	Qwen3.5-27B
Parameters	27B
Device	Jetson AGX Orin 32GB Developer Kit
Quantization	Q4_K_M
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	6.43	6.8	>3 s
2	12.8	6.1	>3 s
4	25.3	5.1	>3 s
8	67.1	5.1	>3 s

Images per request	Response (s)	TPS (tok/s)	Target
1	6.43	6.8	>3 s
2	12.3	6.8	>3 s
4	24.5	6.7	>3 s
8	48.9	6.6	>3 s
16	98.6	6.2	>3 s
32	203.2	5.7	>3 s

Qwen3.5-27B (27B parameters), served in Q4_K_M format with llama.cpp on Jetson AGX Orin 32GB Developer Kit, starts answering a single camera that sends one 720p image after 6.43 s, then writes 6.8 tok/s.

With 32 images per camera at 720p, the answer starts after 203.23 s. With one 1080p image, after 15.94 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.