Profile Job Results
Jobs
jgkyxdk2p
Results Ready
Name
bisenet_w8a8
Target Device
- RB5 (Proxy)
- Android 12
- Qualcomm® QCS8250
Creator
ai-hub-support@qti.qualcomm.com
Target Model
Input Specs
image: uint8[1, 720, 960, 3]Completion Time
1/11/2026, 7:46:11 AM
Options
--qairt_version latestVersions
- TensorFlow Lite: 2.17.0
- Android: 11 (RP1A.200720.012)
- AI Hub: aihub-2026.01.05.0
Estimated Inference Time
166 ms
Estimated Peak Memory Usage
37 ‑ 94 MB
Compute Units
GPU
69
| Stage | Time | Memory |
|---|---|---|
First App Load | 1.61 min | 181‑183 MB |
Subsequent App Load | 217 ms | 98‑100 MB |
Inference | 166 ms | 37‑94 MB |
| TensorFlow Lite | Value |
|---|---|
| number_of_threads | 4 |
| GPUv2 Delegate | Value |
|---|---|
| inference_preference | TFLITE_GPU_INFERENCE_PREFERENCE_SUSTAINED_SPEED |
| inference_priority1 | TFLITE_GPU_INFERENCE_PRIORITY_MIN_LATENCY |
| inference_priority2 | TFLITE_GPU_INFERENCE_PRIORITY_MIN_MEMORY_USAGE |
| inference_priority3 | TFLITE_GPU_INFERENCE_PRIORITY_MAX_PRECISION |
Sign up to run this model on a hosted Qualcomm® device!
Run on device