Profile Job Results
Jobs
jgl122v25
Results Ready
Name
detectron2_detection_roi_head
Target Device
- QCS8550 (Proxy)
- Android 12
- Qualcomm® Dragonwing™ QCS8550
Creator
ai-hub-support@qti.qualcomm.com
Input Specs
features: float32[1, 1024, 50, 50]proposals_boxes: float32[1, 200, 4]Completion Time
7/5/2026, 1:16:05 AM
Versions
- ONNX Runtime: 1.26.0
- QAIRT: v2.45.0.260326154327
- Android: 13 (TP1A.220624.014)
- AI Hub: aihub-2026.06.19.2
Estimated Inference Time
65.6 ms
Estimated Peak Memory Usage
10 ‑ 13 MB
Compute Units
NPU
139
| Stage | Time | Memory |
|---|---|---|
First App Load | 1.60 min | 809‑811 MB |
Subsequent App Load | 553 ms | 40‑151 MB |
Inference | 65.6 ms | 10‑13 MB |
| ONNX Runtime | Value |
|---|---|
| execution_mode | SEQUENTIAL |
| intra_op_num_threads | 0 |
| inter_op_num_threads | 0 |
| enable_memory_pattern | false |
| enable_cpu_memory_arena | false |
| graph_optimization_level | ENABLE_ALL |
| QNN Execution Provider | Value |
|---|---|
| htp_performance_mode | "burst" |
| htp_graph_finalization_optimization_mode | "3" |
| enable_htp_fp16_precision | "1" |
| capture_network_visualizations | false |
| context_priority | "normal" |
| offload_graph_io_quantization | "1" |
Sign up to run this model on a hosted Qualcomm® device!
Run on device