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
| Model | Qwen3-VL-4B-Instruct |
|---|---|
| Parameters | 4.4B |
| Device | RTX PRO 6000 |
| Quantization | Q8_0 |
| Image size | 720p |
| Images per camera | 1 |
| Max Cameras (response ≤ 3 s) | 16+ |
| Engine | llama.cpp |
| Cameras | Response (s) | TPS (tok/s) | Status |
|---|---|---|---|
| 1 | 0.25 | 231.4 | PASS |
| 2 | 0.27 | 161.1 | PASS |
| 4 | 0.44 | 140.0 | PASS |
| 8 | 0.70 | 99.8 | PASS |
| 16 | 1.09 | 73.3 | PASS |
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 0.25 s, then writes 231.4 tok/s.
With 5 images per camera at 720p, the answer starts after 0.76 s. With one 2K image, after 0.94 s.
At the default target of an answer starting within 3 s (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.
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 0.25 s, then writes 231.4 tok/s.
With 5 images per camera at 720p, the answer starts after 0.76 s. With one 2K image, after 0.94 s.
At the default target of an answer starting within 3 s (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.