Profile Job Results
Jobs
jp2rjd1mg
Results Ready
Name
mediapipe_face_face_landmark_detector
Target Device
- Dragonwing IQ-9075 EVK
- Qualcomm Linux 1.9
- Qualcomm® Dragonwing™ IQ-9075 | QCS9075
Creator
ai-hub-support@qti.qualcomm.com
Input Specs
image: uint8[1, 192, 192, 3]Completion Time
9/20/2026, 12:14:31 AM
Options
--qairt_version latestVersions
- Lite RT: 1.4.2
- QAIRT: v2.50.0.260828221209
- QNN TfLite Delegate: v2.50.0.260828221209
- Qualcomm Linux: 1.9-ver.1.0
- AI Hub: aihub-2026.09.11.0
Estimated Inference Time
236 μs
Estimated Peak Memory Usage
0 ‑ 3 MB
Compute Units
NPU
106
| Stage | Time | Memory |
|---|---|---|
First App Load | 620 ms | 46‑47 MB |
Subsequent App Load | 203 ms | 5‑6 MB |
Inference | 236 μs | 0‑3 MB |
| Lite RT | 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.optimization_strategy | kHtpOptimizeForInferenceO3 |
| htp_options.useConvHmx | true |
Sign up to run this model on a hosted Qualcomm® device!
Run on device