Profile Job Results
Jobs
j570yjvrg
Results Ready
Name
mediapipe_hand_HandLandmarkDetector
Target Device
- QCS8550 (Proxy)
- Android 12
- Qualcomm® QCS8550
Creator
ai-hub-support@qti.qualcomm.com
Target Model
Input Specs
image: float32[1, 256, 256, 3]Completion Time
4/8/2025, 11:26:28 AM
Versions
- TensorFlow Lite: 2.17.0
- QAIRT: v2.32.0.250228225014_116386
- QNN TfLite Delegate: v2.32.0.250228225014_116386
- Android: 13 (TP1A.220624.014)
- AI Hub: aihub-2025.03.24.0
Estimated Inference Time
1.01 s
Estimated Peak Memory Usage
28 ‑ 65,839 GB
Compute Units
NPU
158
| Stage | Time | Memory |
|---|---|---|
First App Load | 12.7 min | 61,936‑63,053 GB |
Subsequent App Load | 11.9 min | 99,312‑102,277 GB |
Inference | 1.01 s | 28‑65,839 GB |
| TensorFlow Lite | Value |
|---|---|
| number_of_threads | 4 |
| QNN Delegate | Value |
|---|---|
| backend_type | kHtpBackend |
| log_level | kLogLevelWarn |
| htp_options.performance_mode | kHtpBurst |
| htp_options.precision | kHtpFp16 |
| htp_options.useConvHmx | true |
Sign up to run this model on a hosted Qualcomm® device!
Run on device