VLM Inference Benchmark Explorer
Cosmos3-Edge on Jetson AGX Orin 32GB — BF16, vLLM vision-language inference benchmark
Open this configuration in the VLM Inference Benchmark Explorer
| Model | Cosmos3-Edge |
|---|---|
| Parameters | 3.9B |
| Device | Jetson AGX Orin 32GB |
| Quantization | BF16 |
| Image size | 720p |
| Images per camera | 1 |
| Max Cameras (response ≤ 3 s) | 8+ |
| Engine | vLLM |
| Cameras | Response (s) | TPS (tok/s) | Target |
|---|---|---|---|
| 1 | 0.68 | 42.6 | ≤3 s |
| 2 | 0.80 | 35.6 | ≤3 s |
| 4 | 1.60 | 32.0 | ≤3 s |
| 8 | 2.89 | 26.2 | ≤3 s |
| Images per request | Response (s) | TPS (tok/s) | Target |
|---|---|---|---|
| 1 | 0.68 | 42.6 | ≤3 s |
| 3 | 1.76 | 40.5 | ≤3 s |
| 5 | 2.96 | 38.5 | ≤3 s |
Cosmos3-Edge (3.9B parameters), served in BF16 format with vLLM on Jetson AGX Orin 32GB, starts answering a single camera that sends one 720p image after 0.68 s, then writes 42.6 tok/s.
With 5 images per camera at 720p, the answer starts after 2.96 s. With one 2K image, after 3.95 s.
At the default target of an answer starting within 3 s (720p, one image per camera), it keeps up with at least 8 cameras at once: it still met the target at the highest number measured, so the real limit is higher.
Cosmos3-Edge (3.9B parameters), served in BF16 format with vLLM on Jetson AGX Orin 32GB, starts answering a single camera that sends one 720p image after 0.68 s, then writes 42.6 tok/s.
With 5 images per camera at 720p, the answer starts after 2.96 s. With one 2K image, after 3.95 s.
At the default target of an answer starting within 3 s (720p, one image per camera), it keeps up with at least 8 cameras at once: it still met the target at the highest number measured, so the real limit is higher.