VLM Inference Benchmark Explorer

Cosmos-Reason2-2B on Jetson AGX Orin 32GB — Q8_0, llama.cpp vision-language inference benchmark

Open this configuration in the VLM Inference Benchmark Explorer

ModelCosmos-Reason2-2B
Parameters2.4B
DeviceJetson AGX Orin 32GB
QuantizationQ8_0
Image size720p
Images per camera1
Max Cameras (TTFT ≤ 3000 ms)2
Enginellama.cpp
CamerasTTFT (ms)TPS (tok/s)Target
1134844.4≤3000 ms
2182725.8≤3000 ms
4302417.0>3000 ms
8859621.1>3000 ms
Images per requestTTFT (ms)TPS (tok/s)Target
1134844.4≤3000 ms
3377843.8>3000 ms
5626141.5>3000 ms

Cosmos-Reason2-2B (2.4B parameters), served in Q8_0 format with llama.cpp on Jetson AGX Orin 32GB, starts answering a single camera that sends one 720p image after 1,348 ms (time to first token, TTFT), then writes 44.4 tok/s.

With 5 images per camera at 720p, the answer starts after 6,261 ms. With one 2K image, after 9,690 ms.

At the default target of a mean TTFT within 3,000 ms (720p, one image per camera), it keeps up with 2 cameras at once.