VLM Inference Benchmark Explorer

Qwen3-VL-4B-Instruct on RTX PRO 6000 — Q8_0, llama.cpp vision-language inference benchmark

Open this configuration in the VLM Inference Benchmark Explorer

ModelQwen3-VL-4B-Instruct
Parameters4.4B
DeviceRTX PRO 6000
QuantizationQ8_0
Image size720p
Images per camera1
Max Cameras (TTFT ≤ 3000 ms)16+
Enginellama.cpp
CamerasTTFT (ms)TPS (tok/s)Target
1247231.4≤3000 ms
2266161.1≤3000 ms
4441140.0≤3000 ms
869599.8≤3000 ms
16109473.3≤3000 ms
Images per requestTTFT (ms)TPS (tok/s)Target
1247231.4≤3000 ms
3541233.4≤3000 ms
5761223.5≤3000 ms

Qwen3-VL-4B-Instruct (4.4B parameters), served in Q8_0 format with llama.cpp on RTX PRO 6000, starts answering a single camera that sends one 720p image after 247 ms (time to first token, TTFT), then writes 231.4 tok/s.

With 5 images per camera at 720p, the answer starts after 761 ms. With one 2K image, after 944 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.