Profile Job Results
Jobs
jgz7llvkp
Results Ready
Name
zipformer_ZipformerEncoder
Target Device
- QCS8550 (Proxy)
- Android 12
- Qualcomm® QCS8550
Creator
ai-hub-support@qti.qualcomm.com
Target Model
Input Specs
x: float32[1, 71, 80]cached_len_0: int32[2, 1]cached_len_1: int32[4, 1]cached_len_2: int32[3, 1]cached_len_3: int32[2, 1]cached_len_4: int32[4, 1]cached_avg_0: float32[2, 1, 384]cached_avg_1: float32[4, 1, 384]cached_avg_2: float32[3, 1, 384]cached_avg_3: float32[2, 1, 384]cached_avg_4: float32[4, 1, 384]cached_key_0: float32[2, 128, 1, 192]cached_key_1: float32[4, 64, 1, 192]cached_key_2: float32[3, 32, 1, 192]cached_key_3: float32[2, 16, 1, 192]cached_key_4: float32[4, 64, 1, 192]cached_val_0: float32[2, 128, 1, 96]cached_val_1: float32[4, 64, 1, 96]cached_val_2: float32[3, 32, 1, 96]cached_val_3: float32[2, 16, 1, 96]cached_val_4: float32[4, 64, 1, 96]cached_val2_0: float32[2, 128, 1, 96]cached_val2_1: float32[4, 64, 1, 96]cached_val2_2: float32[3, 32, 1, 96]cached_val2_3: float32[2, 16, 1, 96]cached_val2_4: float32[4, 64, 1, 96]cached_conv1_0: float32[2, 1, 384, 30]cached_conv1_1: float32[4, 1, 384, 30]cached_conv1_2: float32[3, 1, 384, 30]cached_conv1_3: float32[2, 1, 384, 30]cached_conv1_4: float32[4, 1, 384, 30]cached_conv2_0: float32[2, 1, 384, 30]cached_conv2_1: float32[4, 1, 384, 30]cached_conv2_2: float32[3, 1, 384, 30]cached_conv2_3: float32[2, 1, 384, 30]cached_conv2_4: float32[4, 1, 384, 30]Completion Time
2/24/2026, 6:53:47 AM
Options
--qairt_version 2.42 --max_profiler_iterations 10Versions
- ONNX Runtime: 1.24.1
- QAIRT: v2.42.0.251225135753_193295
- Android: 13 (TP1A.220624.014)
- AI Hub: aihub-2026.02.14.0
Estimated Inference Time
9.06 ms
Estimated Peak Memory Usage
9 ‑ 10 MB
Compute Units
NPU
2561
| Stage | Time | Memory |
|---|---|---|
First App Load | 364 ms | 151‑153 MB |
Subsequent App Load | 323 ms | 146‑148 MB |
Inference | 9.06 ms | 9‑10 MB |
| ONNX Runtime | Value |
|---|---|
| execution_mode | SEQUENTIAL |
| intra_op_num_threads | 0 |
| inter_op_num_threads | 0 |
| enable_memory_pattern | false |
| enable_cpu_memory_arena | false |
| graph_optimization_level | ENABLE_ALL |
| QNN Execution Provider | Value |
|---|---|
| htp_performance_mode | "burst" |
| htp_graph_finalization_optimization_mode | "3" |
| enable_htp_fp16_precision | "1" |
| capture_network_visualizations | false |
| context_priority | "normal" |
| offload_graph_io_quantization | "1" |
Sign up to run this model on a hosted Qualcomm® device!
Run on device