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
| Model | Qwen3.8-27B |
|---|---|
| Parameters | 27B |
| Device | RTX PRO 6000 |
| Quantization | FP8 |
| Image size | 720p |
| Images per camera | 1 |
| Max Cameras (TTFT ≤ 3000 ms) | 16+ |
| Engine | vLLM |
| Cameras | TTFT (ms) | TPS (tok/s) | Target |
|---|---|---|---|
| 1 | 293 | 50.0 | ≤3000 ms |
| 2 | 372 | 46.1 | ≤3000 ms |
| 4 | 514 | 45.2 | ≤3000 ms |
| 8 | 866 | 44.1 | ≤3000 ms |
| 16 | 1378 | 41.6 | ≤3000 ms |
| Images per request | TTFT (ms) | TPS (tok/s) | Target |
|---|---|---|---|
| 1 | 293 | 50.0 | ≤3000 ms |
| 3 | 787 | 49.4 | ≤3000 ms |
| 5 | 1014 | 49.9 | ≤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.
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.