Profile Job Results
Jobs
jp8m77q85
Results Ready
Name
deeplabv3_resnet50
Target Device
- QCS6490 (Proxy)
- Android 12
- Qualcomm® QCS6490
Creator
ai-hub-support@qti.qualcomm.com
Target Model
Input Specs
image: uint8[1, 520, 520, 3]Completion Time
7/20/2025, 1:12:43 AM
Options
--compute_unit gpuVersions
- TensorFlow Lite: 2.17.0
- Android: 12 (SP1A.210812.016)
- AI Hub: aihub-2025.07.12.0
Estimated Inference Time
1.39 s
Estimated Peak Memory Usage
4 ‑ 33 MB
Compute Units
GPU
100
CPU
2
| Stage | Time | Memory |
|---|---|---|
First App Load | 3.43 min | 486‑498 MB |
Subsequent App Load | 3.40 min | 490‑519 MB |
Inference | 1.39 s | 4‑33 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