VLM Inference Benchmark Explorer

Qwen3.8-27B on RTX PRO 6000 — FP8, vLLM vision-language inference benchmark

Open this configuration in the VLM Inference Benchmark Explorer

ModelQwen3.8-27B
Parameters27B
DeviceRTX PRO 6000
QuantizationFP8
Image size720p
Images per camera1
Max Cameras (TTFT ≤ 3000 ms)16+
EnginevLLM
CamerasTTFT (ms)TPS (tok/s)Target
129350.0≤3000 ms
237246.1≤3000 ms
451445.2≤3000 ms
886644.1≤3000 ms
16137841.6≤3000 ms
Images per requestTTFT (ms)TPS (tok/s)Target
129350.0≤3000 ms
378749.4≤3000 ms
5101449.9≤3000 ms

Qwen3.8-27B (27B parameters), served in FP8 format with vLLM on RTX PRO 6000, starts answering a single camera that sends one 720p image after 293 ms (time to first token, TTFT), then writes 50 tok/s.

With 5 images per camera at 720p, the answer starts after 1,014 ms. With one 2K image, after 906 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.