VLM Inference Benchmark Explorer
Qwen3.5-4B on Jetson AGX Orin 32GB — Q4_K_M, llama.cpp vision-language inference benchmark
Open this configuration in the VLM Inference Benchmark Explorer
| Model | Qwen3.5-4B |
|---|---|
| Parameters | 4.7B |
| Device | Jetson AGX Orin 32GB |
| Quantization | Q4_K_M |
| Image size | 720p |
| Images per camera | 1 |
| Max Cameras (TTFT ≤ 3000 ms) | 1 |
| Engine | llama.cpp |
| Cameras | TTFT (ms) | TPS (tok/s) | Target |
|---|---|---|---|
| 1 | 2210 | 25.3 | ≤3000 ms |
| 2 | 3574 | 20.7 | >3000 ms |
| 4 | 7150 | 13.1 | >3000 ms |
| 8 | 14740 | 13.0 | >3000 ms |
| Images per request | TTFT (ms) | TPS (tok/s) | Target |
|---|---|---|---|
| 1 | 2210 | 25.3 | ≤3000 ms |
| 3 | 6312 | 25.1 | >3000 ms |
| 5 | 10385 | 25.0 | >3000 ms |
Qwen3.5-4B (4.7B parameters), served in Q4_K_M format with llama.cpp on Jetson AGX Orin 32GB, starts answering a single camera that sends one 720p image after 2,210 ms (time to first token, TTFT), then writes 25.3 tok/s.
With 5 images per camera at 720p, the answer starts after 10,385 ms. With one 2K image, after 12,483 ms.
At the default target of a mean TTFT within 3,000 ms (720p, one image per camera), it keeps up with one camera at a time.
Qwen3.5-4B (4.7B parameters), served in Q4_K_M format with llama.cpp on Jetson AGX Orin 32GB, starts answering a single camera that sends one 720p image after 2,210 ms (time to first token, TTFT), then writes 25.3 tok/s.
With 5 images per camera at 720p, the answer starts after 10,385 ms. With one 2K image, after 12,483 ms.
At the default target of a mean TTFT within 3,000 ms (720p, one image per camera), it keeps up with one camera at a time.