VLM Inference Benchmark Explorer

Cosmos-Reason1-7B on Jetson AGX Thor — BF16, vLLM vision-language inference benchmark

Open this configuration in the VLM Inference Benchmark Explorer

ModelCosmos-Reason1-7B
Parameters8.3B
DeviceJetson AGX Thor
QuantizationBF16
Image size720p
Images per camera1
Max Cameras (TTFT ≤ 3000 ms)8
EnginevLLM
CamerasTTFT (ms)TPS (tok/s)Target
158216.4≤3000 ms
276415.9≤3000 ms
4131515.3≤3000 ms
8189613.6≤3000 ms
16333211.9>3000 ms
Images per requestTTFT (ms)TPS (tok/s)Target
158216.4≤3000 ms
2101216.2≤3000 ms
4197516.1≤3000 ms
8409516.0>3000 ms
16891815.2>3000 ms
322132014.5>3000 ms
645610012.8>3000 ms
128——Failed

Cosmos-Reason1-7B (8.3B parameters), served in BF16 format with vLLM on Jetson AGX Thor, starts answering a single camera that sends one 720p image after 582 ms (time to first token, TTFT), then writes 16.4 tok/s.

With 64 images per camera at 720p, the answer starts after 56,100 ms. With one 2K image, after 2,297 ms.

At the default target of a mean TTFT within 3,000 ms (720p, one image per camera), it keeps up with 8 cameras at once.