Profile Job Results
Jobs
jgk6jmnw5
Results Ready
Name
detectron2_panoptic_seg_roi_head
Target Device
- Dragonwing IQ-9075 EVK
- Qualcomm Linux 1.9
- Qualcomm® Dragonwing™ IQ-9075 | QCS9075
Creator
ai-hub-support@qti.qualcomm.com
Input Specs
feature_p2: float32[1, 256, 200, 200]feature_p3: float32[1, 256, 100, 100]feature_p4: float32[1, 256, 50, 50]feature_p5: float32[1, 256, 25, 25]proposals: float32[200, 4]Completion Time
10/20/2026, 8:21:53 AM
Options
--qairt_version 2.50Versions
- ONNX Runtime: 1.30.0
- QNN Execution Provider: 2.5.0
- QAIRT: v2.50.0.260828221209
- Qualcomm Linux: 1.9-ver.1.0
- AI Hub: aihub-2026.10.10.0
Estimated Inference Time
79.8 ms
Estimated Peak Memory Usage
52 ‑ 94 MB
Compute Units
NPU
383
CPU
8
| Stage | Time | Memory |
|---|---|---|
First App Load | 2.93 min | 804 MB |
Subsequent App Load | 559 ms | 40 MB |
Inference | 79.8 ms | 52‑94 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