VLM Inference Benchmark Explorer

Cosmos-Reason2-8B on RTX PRO 6000 — BF16, vLLM vision-language inference benchmark

Open this configuration in the VLM Inference Benchmark Explorer

ModelCosmos-Reason2-8B
Parameters8.8B
DeviceRTX PRO 6000
QuantizationBF16
Image size720p
Images per camera1
Max Cameras (TTFT ≤ 3000 ms)16+
EnginevLLM
CamerasTTFT (ms)TPS (tok/s)Target
114892.5≤3000 ms
220389.2≤3000 ms
425880.1≤3000 ms
829467.6≤3000 ms
1654554.8≤3000 ms
Images per requestTTFT (ms)TPS (tok/s)Target
114892.5≤3000 ms
344692.3≤3000 ms
575092.4≤3000 ms

Cosmos-Reason2-8B (8.8B parameters), served in BF16 format with vLLM on RTX PRO 6000, starts answering a single camera that sends one 720p image after 148 ms (time to first token, TTFT), then writes 92.5 tok/s.

With 5 images per camera at 720p, the answer starts after 750 ms. With one 2K image, after 680 ms.

At the default target of a mean TTFT within 3,000 ms (720p, one image per camera), it keeps up with at least 16 cameras at once: it still met the target at the highest number measured, so the real limit is higher.