VLM Inference Benchmark Explorer
Eagle2.5-8B on RTX PRO 6000 — BF16, vLLM vision-language inference benchmark
Open this configuration in the VLM Inference Benchmark Explorer
| Model | Eagle2.5-8B |
|---|---|
| Parameters | 8.1B |
| Device | RTX PRO 6000 |
| Quantization | BF16 |
| Image size | 720p |
| Images per camera | 1 |
| Max Cameras (TTFT ≤ 3000 ms) | 16+ |
| Engine | vLLM |
| Cameras | TTFT (ms) | TPS (tok/s) | Target |
|---|---|---|---|
| 1 | 258 | 105.0 | ≤3000 ms |
| 2 | 296 | 154.3 | ≤3000 ms |
| 4 | 426 | 81.2 | ≤3000 ms |
| 8 | 599 | 53.9 | ≤3000 ms |
| 16 | 1321 | 29.5 | ≤3000 ms |
| Images per request | TTFT (ms) | TPS (tok/s) | Target |
|---|---|---|---|
| 1 | 258 | 105.0 | ≤3000 ms |
| 3 | 778 | 124.5 | ≤3000 ms |
| 5 | 1352 | 105.9 | ≤3000 ms |
Eagle2.5-8B (8.1B parameters), served in BF16 format with vLLM on RTX PRO 6000, starts answering a single camera that sends one 720p image after 258 ms (time to first token, TTFT), then writes 105 tok/s.
With 5 images per camera at 720p, the answer starts after 1,352 ms. With one 2K image, after 418 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.
Eagle2.5-8B (8.1B parameters), served in BF16 format with vLLM on RTX PRO 6000, starts answering a single camera that sends one 720p image after 258 ms (time to first token, TTFT), then writes 105 tok/s.
With 5 images per camera at 720p, the answer starts after 1,352 ms. With one 2K image, after 418 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.