Profile Job Results
Jobs
jp8mqlmk5
Results Ready
Name
mediapipe_face_FaceLandmarkDetector
Target Device
- SA8255 (Proxy)
- Android 13
- Qualcomm® SA8255P
Creator
ai-hub-support@qti.qualcomm.com
Target Model
Input Specs
image: uint8[1, 192, 192, 3]Completion Time
6/28/2025, 2:00:25 PM
Versions
- TensorFlow Lite: 2.17.0
- QAIRT: v2.34.2.250528164111_119506
- QNN TfLite Delegate: v2.34.2.250528164111_119506
- Android: 13 (TP1A.220624.014)
- AI Hub: aihub-2025.06.12.0
Estimated Inference Time
155 μs
Estimated Peak Memory Usage
0 ‑ 12 MB
Compute Units
NPU
108
| Stage | Time | Memory |
|---|---|---|
First App Load | 598 ms | 25‑26 MB |
Subsequent App Load | 552 ms | 0‑4 GB |
Inference | 155 μs | 0‑12 MB |
| 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