Profile Job Results
Jobs
jgklo8jn5
Results Ready
Name
distil_bert_base_uncased_hf
Target Device
- QCS8450 (Proxy)
- Android 13
- Qualcomm® QCS8450
Creator
ai-hub-support@qti.qualcomm.com
Target Model
Input Specs
input_tokens: int32[1, 384]attention_masks: float32[1, 384]mask_indices: int32[1]Completion Time
4/5/2026, 9:52:13 AM
Versions
- LiteRT: 1.4.2
- QAIRT: v2.43.0.260127150333_193827
- QNN TfLite Delegate: v2.43.0.260127150333_193827
- Android: 13 (TP1A.220624.014)
- AI Hub: aihub-2026.03.29.0
Estimated Inference Time
21.3 ms
Estimated Peak Memory Usage
0 ‑ 403 MB
Compute Units
NPU
290
| Stage | Time | Memory |
|---|---|---|
First App Load | 6.83 s | 728‑736 MB |
Subsequent App Load | 358 ms | 175‑580 MB |
Inference | 21.3 ms | 0‑403 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.optimization_strategy | kHtpOptimizeForInferenceO3 |
| htp_options.useConvHmx | true |
Sign up to run this model on a hosted Qualcomm® device!
Run on device